process_crystal_graph.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import sys"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.7.14 (default, Feb 15 2018, 20:22:28) \n",
      "[GCC 4.2.1 Compatible Apple LLVM 9.0.0 (clang-900.0.39.2)]\n"
     ]
    }
   ],
   "source": [
    "print sys.version"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import networkx as nx\n",
    "from tqdm import tqdm_notebook"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from cgnode2vec import GraphNV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python2.7/site-packages/pymatgen/__init__.py:87: UserWarning: \n",
      "Pymatgen will drop Py2k support from v2019.1.1. Pls consult the documentation\n",
      "at https://www.pymatgen.org for more details.\n",
      "  at https://www.pymatgen.org for more details.\"\"\")\n"
     ]
    }
   ],
   "source": [
    "from utils import get_xyz_data, get_size\n",
    "from utils import get_factor, get_crytal_graph, convert_to_red, get_shortest_distances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from properties import get_element_properties\n",
    "from properties import ElementExtended\n",
    "from dataload import load_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "DATA_DIR = './data'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Get base train and test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "train, test = load_features(DATA_DIR, with_ext=False, with_geo=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>spacegroup</th>\n",
       "      <th>natoms</th>\n",
       "      <th>al</th>\n",
       "      <th>ga</th>\n",
       "      <th>in</th>\n",
       "      <th>a</th>\n",
       "      <th>b</th>\n",
       "      <th>c</th>\n",
       "      <th>alpha</th>\n",
       "      <th>beta</th>\n",
       "      <th>gamma</th>\n",
       "      <th>E0</th>\n",
       "      <th>bandgap</th>\n",
       "      <th>spacegroup_natoms</th>\n",
       "      <th>cellvol</th>\n",
       "      <th>atom_density</th>\n",
       "      <th>alpha_r</th>\n",
       "      <th>beta_r</th>\n",
       "      <th>gamma_r</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.6250</td>\n",
       "      <td>0.3750</td>\n",
       "      <td>0.000</td>\n",
       "      <td>9.9523</td>\n",
       "      <td>8.5513</td>\n",
       "      <td>9.1775</td>\n",
       "      <td>90.0026</td>\n",
       "      <td>90.0023</td>\n",
       "      <td>90.0017</td>\n",
       "      <td>0.0680</td>\n",
       "      <td>3.4387</td>\n",
       "      <td>33_80</td>\n",
       "      <td>781.052081</td>\n",
       "      <td>0.102426</td>\n",
       "      <td>1.570842</td>\n",
       "      <td>1.570836</td>\n",
       "      <td>1.570826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>194</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.6250</td>\n",
       "      <td>0.3750</td>\n",
       "      <td>0.000</td>\n",
       "      <td>6.1840</td>\n",
       "      <td>6.1838</td>\n",
       "      <td>23.6287</td>\n",
       "      <td>90.0186</td>\n",
       "      <td>89.9980</td>\n",
       "      <td>120.0025</td>\n",
       "      <td>0.2490</td>\n",
       "      <td>2.9210</td>\n",
       "      <td>194_80</td>\n",
       "      <td>782.500110</td>\n",
       "      <td>0.102236</td>\n",
       "      <td>1.571121</td>\n",
       "      <td>1.570761</td>\n",
       "      <td>2.094439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>227</td>\n",
       "      <td>40.0</td>\n",
       "      <td>0.8125</td>\n",
       "      <td>0.1875</td>\n",
       "      <td>0.000</td>\n",
       "      <td>9.7510</td>\n",
       "      <td>5.6595</td>\n",
       "      <td>13.9630</td>\n",
       "      <td>90.9688</td>\n",
       "      <td>91.1228</td>\n",
       "      <td>30.5185</td>\n",
       "      <td>0.1821</td>\n",
       "      <td>2.7438</td>\n",
       "      <td>227_40</td>\n",
       "      <td>391.227531</td>\n",
       "      <td>0.102242</td>\n",
       "      <td>1.587705</td>\n",
       "      <td>1.590393</td>\n",
       "      <td>0.532648</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>167</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.7500</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.250</td>\n",
       "      <td>5.0036</td>\n",
       "      <td>5.0034</td>\n",
       "      <td>13.5318</td>\n",
       "      <td>89.9888</td>\n",
       "      <td>90.0119</td>\n",
       "      <td>120.0017</td>\n",
       "      <td>0.2172</td>\n",
       "      <td>3.3492</td>\n",
       "      <td>167_30</td>\n",
       "      <td>293.377334</td>\n",
       "      <td>0.102257</td>\n",
       "      <td>1.570601</td>\n",
       "      <td>1.571004</td>\n",
       "      <td>2.094425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>194</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.6250</td>\n",
       "      <td>0.375</td>\n",
       "      <td>6.6614</td>\n",
       "      <td>6.6612</td>\n",
       "      <td>24.5813</td>\n",
       "      <td>89.9960</td>\n",
       "      <td>90.0006</td>\n",
       "      <td>119.9893</td>\n",
       "      <td>0.0505</td>\n",
       "      <td>1.3793</td>\n",
       "      <td>194_80</td>\n",
       "      <td>944.713843</td>\n",
       "      <td>0.084682</td>\n",
       "      <td>1.570727</td>\n",
       "      <td>1.570807</td>\n",
       "      <td>2.094208</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  spacegroup  natoms      al      ga     in       a       b        c  \\\n",
       "0   1          33    80.0  0.6250  0.3750  0.000  9.9523  8.5513   9.1775   \n",
       "1   2         194    80.0  0.6250  0.3750  0.000  6.1840  6.1838  23.6287   \n",
       "2   3         227    40.0  0.8125  0.1875  0.000  9.7510  5.6595  13.9630   \n",
       "3   4         167    30.0  0.7500  0.0000  0.250  5.0036  5.0034  13.5318   \n",
       "4   5         194    80.0  0.0000  0.6250  0.375  6.6614  6.6612  24.5813   \n",
       "\n",
       "     alpha     beta     gamma      E0  bandgap spacegroup_natoms     cellvol  \\\n",
       "0  90.0026  90.0023   90.0017  0.0680   3.4387             33_80  781.052081   \n",
       "1  90.0186  89.9980  120.0025  0.2490   2.9210            194_80  782.500110   \n",
       "2  90.9688  91.1228   30.5185  0.1821   2.7438            227_40  391.227531   \n",
       "3  89.9888  90.0119  120.0017  0.2172   3.3492            167_30  293.377334   \n",
       "4  89.9960  90.0006  119.9893  0.0505   1.3793            194_80  944.713843   \n",
       "\n",
       "   atom_density   alpha_r    beta_r   gamma_r  \n",
       "0      0.102426  1.570842  1.570836  1.570826  \n",
       "1      0.102236  1.571121  1.570761  2.094439  \n",
       "2      0.102242  1.587705  1.590393  0.532648  \n",
       "3      0.102257  1.570601  1.571004  2.094425  \n",
       "4      0.084682  1.570727  1.570807  2.094208  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>spacegroup</th>\n",
       "      <th>natoms</th>\n",
       "      <th>al</th>\n",
       "      <th>ga</th>\n",
       "      <th>in</th>\n",
       "      <th>a</th>\n",
       "      <th>b</th>\n",
       "      <th>c</th>\n",
       "      <th>alpha</th>\n",
       "      <th>beta</th>\n",
       "      <th>gamma</th>\n",
       "      <th>spacegroup_natoms</th>\n",
       "      <th>cellvol</th>\n",
       "      <th>atom_density</th>\n",
       "      <th>alpha_r</th>\n",
       "      <th>beta_r</th>\n",
       "      <th>gamma_r</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.1875</td>\n",
       "      <td>0.4688</td>\n",
       "      <td>0.3438</td>\n",
       "      <td>10.5381</td>\n",
       "      <td>9.0141</td>\n",
       "      <td>9.6361</td>\n",
       "      <td>89.9997</td>\n",
       "      <td>90.0003</td>\n",
       "      <td>90.0006</td>\n",
       "      <td>33_80</td>\n",
       "      <td>915.347470</td>\n",
       "      <td>0.087399</td>\n",
       "      <td>1.570791</td>\n",
       "      <td>1.570802</td>\n",
       "      <td>1.570807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>33</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.7500</td>\n",
       "      <td>0.2500</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>9.8938</td>\n",
       "      <td>8.5014</td>\n",
       "      <td>9.1298</td>\n",
       "      <td>90.0038</td>\n",
       "      <td>90.0023</td>\n",
       "      <td>90.0015</td>\n",
       "      <td>33_80</td>\n",
       "      <td>767.917987</td>\n",
       "      <td>0.104178</td>\n",
       "      <td>1.570863</td>\n",
       "      <td>1.570836</td>\n",
       "      <td>1.570823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>167</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.6667</td>\n",
       "      <td>0.1667</td>\n",
       "      <td>0.1667</td>\n",
       "      <td>4.9811</td>\n",
       "      <td>4.9808</td>\n",
       "      <td>13.4799</td>\n",
       "      <td>89.9900</td>\n",
       "      <td>90.0109</td>\n",
       "      <td>120.0014</td>\n",
       "      <td>167_30</td>\n",
       "      <td>289.624655</td>\n",
       "      <td>0.103582</td>\n",
       "      <td>1.570622</td>\n",
       "      <td>1.570987</td>\n",
       "      <td>2.094420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>12</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.5625</td>\n",
       "      <td>0.4375</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>24.3370</td>\n",
       "      <td>6.0091</td>\n",
       "      <td>5.7620</td>\n",
       "      <td>89.9995</td>\n",
       "      <td>103.8581</td>\n",
       "      <td>90.0002</td>\n",
       "      <td>12_80</td>\n",
       "      <td>818.126773</td>\n",
       "      <td>0.097784</td>\n",
       "      <td>1.570788</td>\n",
       "      <td>1.812666</td>\n",
       "      <td>1.570800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>80.0</td>\n",
       "      <td>0.1875</td>\n",
       "      <td>0.5000</td>\n",
       "      <td>0.3125</td>\n",
       "      <td>24.6443</td>\n",
       "      <td>6.2906</td>\n",
       "      <td>6.1589</td>\n",
       "      <td>90.0000</td>\n",
       "      <td>104.5929</td>\n",
       "      <td>90.0001</td>\n",
       "      <td>12_80</td>\n",
       "      <td>923.997043</td>\n",
       "      <td>0.086580</td>\n",
       "      <td>1.570796</td>\n",
       "      <td>1.825490</td>\n",
       "      <td>1.570798</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  spacegroup  natoms      al      ga      in        a       b        c  \\\n",
       "0   1          33    80.0  0.1875  0.4688  0.3438  10.5381  9.0141   9.6361   \n",
       "1   2          33    80.0  0.7500  0.2500  0.0000   9.8938  8.5014   9.1298   \n",
       "2   3         167    30.0  0.6667  0.1667  0.1667   4.9811  4.9808  13.4799   \n",
       "3   4          12    80.0  0.5625  0.4375  0.0000  24.3370  6.0091   5.7620   \n",
       "4   5          12    80.0  0.1875  0.5000  0.3125  24.6443  6.2906   6.1589   \n",
       "\n",
       "     alpha      beta     gamma spacegroup_natoms     cellvol  atom_density  \\\n",
       "0  89.9997   90.0003   90.0006             33_80  915.347470      0.087399   \n",
       "1  90.0038   90.0023   90.0015             33_80  767.917987      0.104178   \n",
       "2  89.9900   90.0109  120.0014            167_30  289.624655      0.103582   \n",
       "3  89.9995  103.8581   90.0002             12_80  818.126773      0.097784   \n",
       "4  90.0000  104.5929   90.0001             12_80  923.997043      0.086580   \n",
       "\n",
       "    alpha_r    beta_r   gamma_r  \n",
       "0  1.570791  1.570802  1.570807  \n",
       "1  1.570863  1.570836  1.570823  \n",
       "2  1.570622  1.570987  2.094420  \n",
       "3  1.570788  1.812666  1.570800  \n",
       "4  1.570796  1.825490  1.570798  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Get element properties from files (as check)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "elem_properties = get_element_properties(DATA_DIR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>element</th>\n",
       "      <th>ea</th>\n",
       "      <th>ip</th>\n",
       "      <th>rs_max</th>\n",
       "      <th>rp_max</th>\n",
       "      <th>rd_max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>In</td>\n",
       "      <td>-0.312500</td>\n",
       "      <td>-5.537400</td>\n",
       "      <td>1.09</td>\n",
       "      <td>1.390000</td>\n",
       "      <td>1.940000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Ga</td>\n",
       "      <td>-0.108100</td>\n",
       "      <td>-5.818200</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1.330000</td>\n",
       "      <td>2.160000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Al</td>\n",
       "      <td>-0.256300</td>\n",
       "      <td>-5.780000</td>\n",
       "      <td>1.13</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>3.110000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>O</td>\n",
       "      <td>-0.225633</td>\n",
       "      <td>-5.711867</td>\n",
       "      <td>1.07</td>\n",
       "      <td>1.406667</td>\n",
       "      <td>2.403333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  element        ea        ip  rs_max    rp_max    rd_max\n",
       "0      In -0.312500 -5.537400    1.09  1.390000  1.940000\n",
       "1      Ga -0.108100 -5.818200    0.99  1.330000  2.160000\n",
       "2      Al -0.256300 -5.780000    1.13  1.500000  3.110000\n",
       "3       O -0.225633 -5.711867    1.07  1.406667  2.403333"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "elem_properties"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Process an example file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "idx = 1\n",
    "fname = os.path.join(DATA_DIR, '{}', '{}', 'geometry.xyz').format('train', idx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos, lat, natoms = get_xyz_data(fname)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "      <th>element</th>\n",
       "      <th>electroneg</th>\n",
       "      <th>atomic_mass</th>\n",
       "      <th>r_ionic</th>\n",
       "      <th>ea</th>\n",
       "      <th>ip</th>\n",
       "      <th>rs_max</th>\n",
       "      <th>rp_max</th>\n",
       "      <th>rd_max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.608888</td>\n",
       "      <td>7.276416</td>\n",
       "      <td>6.383155</td>\n",
       "      <td>Ga</td>\n",
       "      <td>1.81</td>\n",
       "      <td>69.723000</td>\n",
       "      <td>0.760</td>\n",
       "      <td>-0.1081</td>\n",
       "      <td>-5.8182</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1.33</td>\n",
       "      <td>2.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6.584932</td>\n",
       "      <td>7.252792</td>\n",
       "      <td>6.365826</td>\n",
       "      <td>Al</td>\n",
       "      <td>1.61</td>\n",
       "      <td>26.981539</td>\n",
       "      <td>0.675</td>\n",
       "      <td>-0.2563</td>\n",
       "      <td>-5.7800</td>\n",
       "      <td>1.13</td>\n",
       "      <td>1.50</td>\n",
       "      <td>3.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.435758</td>\n",
       "      <td>1.259776</td>\n",
       "      <td>1.794613</td>\n",
       "      <td>Al</td>\n",
       "      <td>1.61</td>\n",
       "      <td>26.981539</td>\n",
       "      <td>0.675</td>\n",
       "      <td>-0.2563</td>\n",
       "      <td>-5.7800</td>\n",
       "      <td>1.13</td>\n",
       "      <td>1.50</td>\n",
       "      <td>3.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8.411802</td>\n",
       "      <td>1.236152</td>\n",
       "      <td>1.777284</td>\n",
       "      <td>Ga</td>\n",
       "      <td>1.81</td>\n",
       "      <td>69.723000</td>\n",
       "      <td>0.760</td>\n",
       "      <td>-0.1081</td>\n",
       "      <td>-5.8182</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1.33</td>\n",
       "      <td>2.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.955880</td>\n",
       "      <td>2.998925</td>\n",
       "      <td>1.801351</td>\n",
       "      <td>Ga</td>\n",
       "      <td>1.81</td>\n",
       "      <td>69.723000</td>\n",
       "      <td>0.760</td>\n",
       "      <td>-0.1081</td>\n",
       "      <td>-5.8182</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1.33</td>\n",
       "      <td>2.16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          x         y         z element  electroneg  atomic_mass  r_ionic  \\\n",
       "0  1.608888  7.276416  6.383155      Ga        1.81    69.723000    0.760   \n",
       "1  6.584932  7.252792  6.365826      Al        1.61    26.981539    0.675   \n",
       "2  3.435758  1.259776  1.794613      Al        1.61    26.981539    0.675   \n",
       "3  8.411802  1.236152  1.777284      Ga        1.81    69.723000    0.760   \n",
       "4  0.955880  2.998925  1.801351      Ga        1.81    69.723000    0.760   \n",
       "\n",
       "       ea      ip  rs_max  rp_max  rd_max  \n",
       "0 -0.1081 -5.8182    0.99    1.33    2.16  \n",
       "1 -0.2563 -5.7800    1.13    1.50    3.11  \n",
       "2 -0.2563 -5.7800    1.13    1.50    3.11  \n",
       "3 -0.1081 -5.8182    0.99    1.33    2.16  \n",
       "4 -0.1081 -5.8182    0.99    1.33    2.16  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 9.95208772e+00, -4.72482400e-02, -3.46579300e-02],\n",
       "       [ 4.03159900e-02,  8.55117485e+00, -9.53377000e-03],\n",
       "       [ 3.16403800e-02,  9.66592000e-03,  9.17743912e+00]])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "element\n",
       "Al    20\n",
       "Ga    12\n",
       "O     48\n",
       "dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "natoms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos_R = pos[['x', 'y', 'z']].as_matrix()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos_red = convert_to_red(pos_R, lat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "      <th>element</th>\n",
       "      <th>electroneg</th>\n",
       "      <th>atomic_mass</th>\n",
       "      <th>r_ionic</th>\n",
       "      <th>ea</th>\n",
       "      <th>ip</th>\n",
       "      <th>rs_max</th>\n",
       "      <th>rp_max</th>\n",
       "      <th>rd_max</th>\n",
       "      <th>x_red</th>\n",
       "      <th>y_red</th>\n",
       "      <th>z_red</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.608888</td>\n",
       "      <td>7.276416</td>\n",
       "      <td>6.383155</td>\n",
       "      <td>Ga</td>\n",
       "      <td>1.81</td>\n",
       "      <td>69.723000</td>\n",
       "      <td>0.760</td>\n",
       "      <td>-0.1081</td>\n",
       "      <td>-5.8182</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1.33</td>\n",
       "      <td>2.16</td>\n",
       "      <td>0.156</td>\n",
       "      <td>0.851</td>\n",
       "      <td>0.697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6.584932</td>\n",
       "      <td>7.252792</td>\n",
       "      <td>6.365826</td>\n",
       "      <td>Al</td>\n",
       "      <td>1.61</td>\n",
       "      <td>26.981539</td>\n",
       "      <td>0.675</td>\n",
       "      <td>-0.2563</td>\n",
       "      <td>-5.7800</td>\n",
       "      <td>1.13</td>\n",
       "      <td>1.50</td>\n",
       "      <td>3.11</td>\n",
       "      <td>0.656</td>\n",
       "      <td>0.851</td>\n",
       "      <td>0.697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.435758</td>\n",
       "      <td>1.259776</td>\n",
       "      <td>1.794613</td>\n",
       "      <td>Al</td>\n",
       "      <td>1.61</td>\n",
       "      <td>26.981539</td>\n",
       "      <td>0.675</td>\n",
       "      <td>-0.2563</td>\n",
       "      <td>-5.7800</td>\n",
       "      <td>1.13</td>\n",
       "      <td>1.50</td>\n",
       "      <td>3.11</td>\n",
       "      <td>0.344</td>\n",
       "      <td>0.149</td>\n",
       "      <td>0.197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8.411802</td>\n",
       "      <td>1.236152</td>\n",
       "      <td>1.777284</td>\n",
       "      <td>Ga</td>\n",
       "      <td>1.81</td>\n",
       "      <td>69.723000</td>\n",
       "      <td>0.760</td>\n",
       "      <td>-0.1081</td>\n",
       "      <td>-5.8182</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1.33</td>\n",
       "      <td>2.16</td>\n",
       "      <td>0.844</td>\n",
       "      <td>0.149</td>\n",
       "      <td>0.197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.955880</td>\n",
       "      <td>2.998925</td>\n",
       "      <td>1.801351</td>\n",
       "      <td>Ga</td>\n",
       "      <td>1.81</td>\n",
       "      <td>69.723000</td>\n",
       "      <td>0.760</td>\n",
       "      <td>-0.1081</td>\n",
       "      <td>-5.8182</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1.33</td>\n",
       "      <td>2.16</td>\n",
       "      <td>0.094</td>\n",
       "      <td>0.351</td>\n",
       "      <td>0.197</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          x         y         z element  electroneg  atomic_mass  r_ionic  \\\n",
       "0  1.608888  7.276416  6.383155      Ga        1.81    69.723000    0.760   \n",
       "1  6.584932  7.252792  6.365826      Al        1.61    26.981539    0.675   \n",
       "2  3.435758  1.259776  1.794613      Al        1.61    26.981539    0.675   \n",
       "3  8.411802  1.236152  1.777284      Ga        1.81    69.723000    0.760   \n",
       "4  0.955880  2.998925  1.801351      Ga        1.81    69.723000    0.760   \n",
       "\n",
       "       ea      ip  rs_max  rp_max  rd_max  x_red  y_red  z_red  \n",
       "0 -0.1081 -5.8182    0.99    1.33    2.16  0.156  0.851  0.697  \n",
       "1 -0.2563 -5.7800    1.13    1.50    3.11  0.656  0.851  0.697  \n",
       "2 -0.2563 -5.7800    1.13    1.50    3.11  0.344  0.149  0.197  \n",
       "3 -0.1081 -5.8182    0.99    1.33    2.16  0.844  0.149  0.197  \n",
       "4 -0.1081 -5.8182    0.99    1.33    2.16  0.094  0.351  0.197  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos_all = pos.join(pd.DataFrame(pos_red, columns=['x_red', 'y_red', 'z_red']))\n",
    "pos_all.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "dists, Rij_min = get_shortest_distances(pos_red, lat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAAGUCAYAAAAI+TQuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzsnXucTfX6xz/fmTFjxv2SEKESE0lunW4U6SqKREXFkXTvdPpVcuJ0UZ3qnG6nktM9kqgklVIRXYkQBikklyR34zKX7++PGbPX5zNm77nsMWPmeb9e87Ifa++1vnuttffa6/k8n+frvPcwDMMwjGgTU9IDMAzDMMomdoExDMMwigW7wBiGYRjFgl1gDMMwjGLBLjCGYRhGsWAXGMMwDKNYsAuMYRiGUSzYBcYwDMMoFuwCYxiGYRQLcSU9AMMwDCM3aZt+iVqblQq1j3LRWldBsDsYwzAMo1iwOxjDMIzSSGZGSY+gyNgdjGEYhlEs2AXGMAyjNOIzo/cXAedcM+fc/MDfdufcrfKcM5xz2wLPGR5pvZYiMwzDKI1kRr4wRAvv/TIArQHAORcLYC2Adw/w1Fne+275Xa/dwRiGYRhBugD42Xu/uqgrsguMYRhGKcT7zKj9OecGO+e+D/wNDrPpvgDG5bHsZOfcAufcR865FpHeg7MZLQ3DMEof+377MWpfzvENjs+XD8Y5Fw9gHYAW3vvfZVlVAJne+53OufMBPOm9bxpufXYHYxiGYeznPADz9OICAN777d77ndmPPwRQwTlXO9zKTOQ3DMMojeSj+qsYuAx5pMecc3UB/O699865Dsi6Qfkz3MrsAmMYhlEaOchGS+dcJQBdAVwb+L8hAOC9HwXgEgDXOefSAewG0NdH0FhMgzEMwyiF7Fs9L3oaTKM2JdKLzO5gDMMwSiMlkyKLKnaBMQzDKI0cRKNlcWFVZIZhGEaxYHcwhmEYpRBvKTLDMAyjWLAUmWEYhmEcGLuDMQzDKI1YiswwDMMoFmxGS8MwDMM4MHYHYxiGURqxFJlhGIZRLFgVmWEYhmEcGLuDMQzDKI1YiswwDMMoFixFZhiGYRgHxu5gDMMwSiHeH/o+GLvAGIZhlEbKgAZjKTLDMAyjWLA7GMMwjNJIGRD57QJjGIZRGinvKTLn3LnOuWXOuRXOubuiNSjDMIxyT2ZG9P5KiELfwTjnYgE8A6ArgN8AzHHOTfbeL8lzY/FH+P2PN/U6lpZVaHo4xTGNGvCLK1fhdXXpT/HYE4aHHW/1DP41cFa/XRT71H2hx/LcjybXpvjCOytTPO3hHRS3qPknxfO38OvP6b2N4s/fqkrxHucoTvA5uw1nD+Xn7pycQnHSCdUo/uwNHmtyjc0Ub9mWRHGrR5Ip3j3hS4r1R9XOtRUornND65zHX969hpbtk98zp529keL4vhdSvOHuDyiu2Yo3fsNMfq/PnZdK8fUf8XsbUYmPuQsMp96gxrQMsbEUZiz+heKv3q1Ocbs26ynevCqR4npn8jHdOmcfxXPW8Pl/Qu3QOVRv4JG0bONrqyiuc0VDiqf9ew/F4xN2U/zcGdsprnByS4qxdy+FMad0pfja3m9Q/FD90Dn1xar6tKz7DbwfZz3F7/uEpr9TvOinOhSffAWPPXMXjy1jM8ffzaxLcWrgIKfG8PnXo/smirfOTaO4Whv+epwziY/5yffytpKGPMkH2ShSiqwDgBXe+18AwDn3JoAeAPK8wBiGYRj5pAykyIpygTkCQPAn6m8ATtInOecGAxgMAC62GmJiKhVhk4ZhGOWEMiDyF3uZsvd+tPe+nfe+nV1cDMMwyg9FuYNZCyCY/G2Q/X95EtRdar+9nJbteK4txZmrf6M4/uYHKU7/7PX8jxS5NZdPx/DFrnOPUN7axfJ1VzWX9/+1k+K2orks3lyL4nMuZc3l4wmsHWRI5rZiQHMBgK4B3eWThzh/ftblvK7UBbyt5Bqc807ZUpPisx9rQvHCO1jTOYYPC+kWAGsuALDxmfk5j/fhMFoWD/5FpprLvjffp7hmK97W5oW8cdVcrhPN5Z+iudy7i4/5838LjW/9C6toWZ3T+KDEtjiK4narF1P8/bx6FJ/ai4/D+ul8TBdtYc2lfUPWImpdFtre+pd+pWX1rmLNZeNY1rrGJ/D52mcv60EVTmZNJ+2bRRQn3Hg9xZlfT6M4qLkAwNB1oXPqWdFcJj/DAvMZorks+In3w2lX8DH9ZiyPvV1HPv9jayZQnConaFIgzdS1O4/7PdFWu1/4B8Xb5qVTrJrLNyM2UNxlCKJLOU+RzQHQ1DnXBFkXlr4ALo/KqAzDMMo7ZSBFVugLjPc+3Tl3I4CPAcQCeMl7vzjCywzDMIxyQpGMlt77DwF8GKWxGIZhGPspA3cwzku+vzjZNbxvzsbU51LlunEUq09m0swjwq67z3+aUqw+mT3Db6BYvS6fvxfSTVSv+Uz0GvW5zN1SMJ+M+lw4a53bJ7N7cej16nP5VHwu6pP58V/sNVGfyye3r6RYfTIJiZyHVs1l9+Q5FAfTxnHV+feLai5fDvyGYvXJjKvIHgfVXDZ9w/l99cls+oG3r16Xax8P5dzVI7PizxoUq88l6fwWFKtPRj0V6nNpKftZfTJ/fh3a7+pzWf8qS53qk0lbwGNRn8u0f7LWoD6ZCqJjqOZSvU8zioM+mRndJtAy9bnM+CW8TybjV36++ly+n8n7UX0yk97k8z/odVHvkuo11aR7sWouf77IWpX6ZKo+/3FUfTC7Z74StS/nxI5Xl4hHx5pdGoZhGMVCUZz8FQHMBJCQvZ6J3vsR0RqYYRhGuaYMpMiKosHsBdDZe7/TOVcBwJfOuY+8999GaWyGYRjll/JcpuyzxJv9hpAK2X9hc4ZB3UV9Lqq5qE/mhcPCazCquahPRjUX9boEdRf1yLSM4HO58K7wPhmI5qI+l86Xcm2/+mTOvSzQT0l8LmcP5Zy2+mTOeYw1F/W5JNdgXSOXT+Z2Xn/Q5wIAleWwBNPakXwu+8B9p9QnE8nn8kxH3hfqk1HNRb0uIwKHWT0yj4vmksvn0oh1DvXJzJm0lWL1uVRvH89jE59Mg4Eh3UV9LvUGNubXqk/mRtZc1OcyPoF1CvXJtD+K33vQ5wIAz0tvsqBPRjUX9bl0vzm8T+bcU1lzUZ+Lai7qk+lxMfcXC3pdkhyfX0nyBR7J59L+Iv66VJ8M71UDKGIVWXbDy7kAjgHwjPf+u6iMyjAMo7xTBlJkRRL5vfcZ3vvWyHLxd3DOtdTnOOcGO+e+d859/+IXC4uyOcMwjPKDz4zeXwkRlQnHvPdbnXPTAZwLYJEsGw1gNADsHn9vTh5AW79MkHb7mhIb9Md0ijWllv7esxRrCi5Sy/11o1flPHbgdMkiSYlpGXN6CqfQzr+Yr9ufvcMlr12llDhzNafUOh7HY586LljSXYXKmN+XEuiKYFLHc7v9pu15+Y413G5fU2JrHuP0SsoWTmvV3cCt4YNl0JoS0xRWpDLkme/yfrsK4T8odR+8gOL096dSXO/ao3n9D4TKbzUltnU178mxFbnlTtMvOaVVrzmneoLt9gFu/QIAyODnx3/NabDfXloXiGKpjHnqI3z+tW/IqZqYth0priDt9yt8wWXKmhK7fS2X8/bZy2mqvf/lz1qwDFrb7Wvrl33zOe10wYW87g/e58+9ljFv/4j3UwY4baq2i05H8Htb/mvoe0BTYlOlfDsJnNqe/D63PtJ2/0ZuCn0H45w7zDlXPftxIrLmhVkarYEZB0Y9MkbZRz0yRuEIXlwOCTIzo/dXQhTlDqYegFezdZgYAG9576dEZ1iGYRjlnHJeRbYQwIlRHIthGIZRhjiorWLSNv2SszEtIx5/209hX3tRR26PoWXM2+46nWJtReN3sFahpcTB0mFtBVP3fM7Haxlzl55beFtSEh13HLfE11LiU1rwe9N2MD49tD4tYa4gx0/b0HTtwHqO/ihKuohFGS1Dnr+Oc+raSqbhMO7nHyyDrnMY73Nt5RJTifUfLUO+ag8v11YynV/i+e1U86nQgVuaaDuXvatDJa9ahqyai7aSqT+4McVaAq3tW7SUODaO94W2g8lYGdIOtIQ5sXL4NjTnisanUx5vvuW/FGsZcs89rItoK5kXhubd7j+2Fn82tIz4pIu4fFtbwcS35nVrGXMN0a60LDp9r7R/kXYucSc2D41NypDTRHPRVjJdLwrf7r/fujHRbRXz0VPRaxVz3s0lkmeNishvGIZR2gleXA4JynuZsmEYhmHkRZEuMM656s65ic65pc65FOfcydEamGEYRrnGfDB4EsBU7/0lzrl4QIrShbHidQkSqd2++ly2NeW8c7WHZ1GsPplNC7g1R0XP3pagdqH59fSUVRR36cn5efW5qE8mc/U6ik9pwaXGXy/m2v+zu3MOPeh10TYzaaK5nC9lzJu/ZB1D2+2nTuJ2+1VYCsjlc1HNZc3IuRRvyAzl85/mdDueq8SeiMxdrCXcX501mxotOd+ecHl3ij8fyI0j4sE58YxPOGferg1rCQmNQvpAx/NZh2j6/M8Ua9uZoG8KOMB0zKIVqOaSkS6/7WLZ7zEtcE5pm5ndO/mYdhUN8LrHWdd4aDxrLjWfvJHi52RK5LR5rId2lXb/gx5iPanP3tBnsePpfK6fdBGP5btJ7LHR1i/qc6mRwZ/zLbKfqp7HJ+y3T/H2Tv4re13Sfwg5KY49kv1DX6xlHU59LtMm8TlS2/Pro04ZSJEVpZtyNQAdAVwNAN77fQD2hXuNYRhGSRG8uBgHh6KkyJoA+APAy865H5xzLzjnKkV6kWEYhpEPykCKrCgXmDgAbQA8570/EcAuAHfpk4K9yGbsCl+KbBiGYWRTBpz8hfbBOOfqAvjWe984Oz4dwF3e+wvyes0rR/TLc2NXLLiPYvXJZC7mNvPqc9n7KWsB6pN5u2YnitXrEtRd1CNz/sWcy1efS0yV8D6ZSD6XmEbhW+4HVRb1uajmoj6ZCx7gvLL6XKo0ZB0kl0+mz2kUa7v/DZn83oM+mXq8y3P5XFRzUZ+Mtvvf+8Zkiud8yr2h1CdzUhvWA9Tr0vEfoZy6emRimzeiWH0u/9zBx1h9MnHx4X0uqrmoTyY+MFW1+lxUc1GfTOUarEOoz+W5v/F+U59M5tyZFGu7/xkf83iCPplR57LOpj6XCg1ZX1SfTIZ4UdTnopqL+mR6DGcdTr0uxx4Z0lXUI+Pks6U+l9oZrLmoT6bnhjei64N59+Ho+WAuvuvQ8sF47zc459Y455p575cB6AJgSfSGZhiGET2CF5dDgvLcKiabmwCMza4g+wXAgKIPyTAMwyjXVWQA4L2fD6BdlMZiGIZhlCEOaquY6gHtQr0ie4bfQLHqHLnnc+H8vfpcVHPptfkLfv6Z7JNxtUPr73YdazAxzf5CcdyZl1P8Spt7wU/gcPoS1osuvJDnonG12JPTdTAbSPbMDfWlSrqyMy177e9cOBEvZRvpPy6juOZprJlMmcj59G6XsKbz6wPzOPacz4+XWbIb3HhMzuPp97HOVSmRc9iJ1Vj/iR/Yj2L/02KKE85kD8+4r1i3eKgx5+tjZHKc06/jNPSMwHww7VuxH2PXt6zJPLKT/Rs3pnHuv96lvF9ufYXf6/3v8/qq3cvvtW4PPua7523MeXxODz6ot73A50+v3aznnHIBv5cbx/FnacubrE/WrMzr+0p0i+TGfIxrgI/bf9uE9uP2+TyWmleyh2bj//h8PL4Z641zlrNOVq3nMRTvmMzne40M9rlsfY0nNcwMTMu99NfaOLF16L1VuIK9dtuHv0LxOS3Yk/O35fw5feQI1sKiTnm/gzEMwzhUCF5cDgkOYiPi4sJ6kRmGYRjFQlF7kd3inFvknFvsnLs1WoMyDMMo95QBH0xRWsW0BHANgA7IahEz1Tk3xXu/Iq/XBHUX9Yp07sE9r1wsX/suvJPzxLnnc+H8qPpcVHPJNZ9McsiLEnNUY1qmmkv69DdQECKNvds1rHvEHMtzxye1CM1zn/ra57Qs3rMvYJ9Wu2fKbXYMP0E1lykT2UcTJy9vVU3mg7mJxzrt/tDyD0Vz6b6bTzfVIbaNGMPLr2xDsZf38lBjLjsduop1uqfP4J5dexewL6Z9q9DjOQs59/9eImssQ6uzRnLYRbytRaPZ73H/Mbyf7vmVvSdPfPoxxTFnnUNxYuKMnMe7v2ItoNduHuvbMtYa4/izk3wZf8HEJvN+3fwC62zJjSlEyioee9tTONW06JuQztHuHtb0Nr/GzoWKVXkse7bzMepyB39WPnuEPysVPGsuJybzWBam8PZbHB3SsioN4yLXXSNfpjiGh4LMNP6sqOZyx1ruQchnbxQoAxpMUe5gkgF8571P9d6nA/gCQM/oDMswDMM41CnKBWYRgNOdc7Wcc0kAzgfQUJ8UbBXz0sJVRdicYRhGOaIM9CIripM/xTn3LwCfIKsP2XwAGQd43mgAowFg5+09Dv2yCMMwjINBGUiRFboXWa4VOfcggN+898/m9Zwd15+XszH1uXz+Hmso6pPZ8CFrNIs35z2fC5Bb9wj6XAAg85dVFFd7YEbOY51LZsIsnq9F6fdvmctGNZtPXuFtL+e5Rqb8j70DOvbXHwvti3g5XN1bc35efTKT/sZak2ouqsmoZvP5W9w7qus97PdY8zS/l4XbQstPb8o92FRz+bzvZxRPFs0mDeE1lyqXn0ixk/cyfZhqLuspjg3s5oQTuB+caiotB/M89Zsm/0Hxg5s5H//Ey6ypZIrmcutYCnH/kby+EQHNRn0ubdrw+0g8lRMHWz/g911zEGsuGSnsJUkRzWZ3Bv/ubHnyRornfs06SHLj0NjT9/JYK1Zlz0z6Xt5WzSuPo3jHBO579kMKb0vnP1LNJvVjPt+DusuWu16lZYt/rkPxqxV5rKq5xMqHL6YCx4dN+yK6vcheGxq9XmRXPnRo9SIDAOdcHe/9RufckcjSX/4S6TWGYRhGPigDPpiiGi3fds7VApAG4Abv/dZILzAMwzDyQRlIkRW1F9np0RqIYRiGUbY4qK1igrqL+lxUc1GfjAPH4eZzAQ7gNdH+YuJ1Ceou6pF5rk54DSaST0Y1F/W5XHgnaws69viAtqA+F9Vc1CfT7ZIqFKvPJZImo5pL0OeSNR5eHvTJRPK5TE5kf4X6ZNols9agPpdnZKzqk1HNRb0unQK9ydQj03JweE3mpXjWXO6uyfl61VzU53L/LN4X6pPpHdBd1Odyqmgu6pOpOag9xepzWfcrnwPqk9n3M2suQZ8LwJoLwD6Z45txPzj1udS5phmPTXwyi1aF97lU6c56p/pkznorb6/L4p953UGPDAA8AkZ9Lo8fy9856pOJOmXgDsZaxRiGYZRGDnKZsnOuunNuonNuqXMuxTl3six3zrmnnHMrnHMLnXNt8lrXfqzZpWEYhgEATwKY6r2/JHuOryRZfh6Aptl/JwF4LvvfPIl4B+Oce8k5t9E5tyjwf49mX+UWOufedc5VD7cOwzAMo2D4TB+1v0g456oB6AjgRQDw3u87QNFWDwCv+Sy+BVDdOVcPYYjog3HOdQSwM3vFLbP/72wAn3vv07PNlvDe3xnpTYyvd0XOxtTrMe1hnt9FNZZFm8P7ZDJ3hO9lFtepgHO6BLhu43SK1SdTIZnz9arvpI77kmLVTd68jX0JqrNUzwgdI91vfnP4PmZT/m8VxdpLbOlWzjOr5pKxjF+vGs0U8ckENZ24Vryf9ES/8UHWDtTnMnxVeI2mbTPWWFTzuXUA6yCqk2zYEBq7+lwU1Wi+/Zo/V6r3OMkNqFflq6fZ86M6SzCpoR6ZKqfzflF958+7WAPUXmIJjlMm6nP5eTafA6rRbPuKP6tBn82nw9j7pD6Vmf/i16rGkrY7vE9m+3j2yajGU+mwfRQHdZK/r+T3pT4X9cWoRhNfhfeD9jZL7Hh1VEWZ1FG3RK1OOWnIk2HH5pxrjSxD/BIAJwCYC+AW7/2uwHOmAHjYe/9ldvwZgDu999/ntd6IdzDe+5kANsv/fZLdfwwAvgXQINcLDcMwjFJBsGVX9t9geUocgDYAnvPen4is7ix3FXW70RD5BwL4KK+FwTf2aWqejZYNwzCMIFEU+b33o7337QJ/o2VrvyGrE8t32fFEZF1wgqwF95tskP1/eVLU+WCGAUgHMDav5wTf2FlJx+T1NMMwDCNIpo/eXwS89xsArHHO7a8j74KsdFmQyQCuzK4m+wuAbd779QhDUeaDuRpANwBdfD4bmgX1A/V6tBXNRXuNRfLJdOnJuVftdVaUOV1Uc1GfzI6XTqFY+5xF8qrsc5yf135j4fZb90dZc1HPTatqrNEEe4UBwAX3sAajPpfOfcP3Kgs3n0yPlvxc7RUWaT6XXqK5aK+yzhF8NnfX5Ndrv7D/DA4tV59L05P4fNReZe13siaTy2NzE29bvSpvJ3K+X/uNtWoW0ibUI/PUWeH7nKnmor6VpPqs96jPpU0//ixpr7KW9+U9n0yXO/h8VJ9KB9FctNdYx+F8DqhPJn0vay7a60y9KcF+YZHmc3lENBfVZE6bcB7FOp9MYsercYhzE4Cx2RVkvwAY4JwbAgDe+1EAPkRW1/wVAFIBDMhrRfsp1AXGOXcugDsAdPLepxZmHYZhGEYYDrLR0ns/H0A7+e9RgeUewA0FWWfEC4xzbhyAMwDUds79BmAEgKEAEgBMc1ndTb/13g8pyIYNwzCMMJQBJ3/U2vXnh/fqXk4b01LkuVtCt8dajpuesopiLUP+7B2+3dWU2rg3uWWKEqnl/r5R/8x5rGXIVQa+QrGm1N6dya1mtAz5khPCt9zPXJxCsZYiT5ZS5OC+0/JvLUP+4AFOG2gZ87zt/PxcrWUUSYPdMonLfzUtVrVfKN2iZczbx/xAcaQyZC1jrh0bvuX+ba9wyk2nRV6yPpSm1TLkpFM5ZaZlzN9/HX5a45HJnI7RMmbs5rFrKfLN8t6DpczLl3OKTMuQU9dxOk5TascdxSk1bfe/aDinYbWM+cvXEykOliLPXsr7RcuYt0/ikn0tQ45L4P2oZcyDH+HjoGmxCpV4rJpSi00InYNahvxlb65l0jLm+l9Pj26Z8hPXRq9M+dbnS6Rdf4m1igl3cSlpwl1cSpqCXFxKmoJcXEqacBeXkqYgF5eSJtzFpaQpyMXFiA7WKsYwDKM0UgZSZIVtFfNP59xa59z87L/zi3eYhmEY5YyDWKZcXOTnDuYVAP8F8Jr8/+Pe+8cKsrFgWkzLkC+8i1M7Wo57/sV8LdQy5EhlzJHeaTAtpiXMqrloGXKkMuZnpN2/liFHKmNOuuw03n6gFPnCO7n1uu63CyO0229VnXPUWsbcrXfeZcgA0O3S7RQHT+ZIZcj/zQxfxhyp3X/33ZzP1zLm/1zNKTotRR5aJ6RPPbSV2+k9EKHV/18qc65fy5jf/oHHomXIqrloGXOlO/pTHCxF1pSYljH/6+Tw7fb3ev4saRmzai7a7j/5MtYzg2XMmhLTMuQuksLVMubj64YvQ67apyWPTcqYHzmCzyktRf53k9D5r1Meq+aiZcgtjub9pmXMfAYYQD4uMN77mc65xsU/FCM/qM/FKPuoz8UoHMGLyyFBPtvsl2aKIvLfmN1N+SXnXI3ITzcMwzDyTRlIkRX2AvMcgKMBtAawHsC/83pisBfZm1t+K+TmDMMwjEONfPlgslNkU/a368/vMmVioF3/OZLb/1zavu9xnI9PkHGePZSfn7mac+KpC3j905dww2ct53W1Q/qApqH2zBUPRAHb7d8Qqd1/S87H75v/K8UVLzkz53Fcx0t5Xb2vRzh++Olwivc4/k2RJvv5/IvDpxF2zOdpEWb/wtpE8LjsmMT+neUprBU0acTbqjWS39u2+94MO5YHfuX3dk8Tzv+v+YlvrBsls9409KfQeK7it4W2d7FGuGkMN2qt8RfWd+J69aT47gGfUjxMphao2Ib1qJTXOB1yZNPQWCudzGNZ8DIfs2OSWUOp1IGfv3Qsr/vpONaDHmnCWtncJXxMj6vHloIqDbmVTMVTGuc8vuV51kKH1+Z9Hiu6R1xF9rXA83urejefE5tHTKT4m1U81sqZvL7xiazhDNoTWn+bR9j75nexHgT5rKx7ainF9QcdSXHSbf+Lqtdk10NXRe3Wo9LQV0vEB1PYVjH1Ak3OLgawKNzzDcMwSprgxeWQoARTW9GisK1izsieoMYDWAXg2mIco2EYhnEIkp8qsssO8N8vFsNYDMMwjP2UgSqyg9qLbMfN3XI29vEE9lNIJhYVZVxdRXP55CH2X5zSgue9STqB1x9zJOdq1S/S7ZrQCLQdCyrEU6g+lYkLwrfbv7gjj019Mn/04FxwfGvO7cYPHJrzOH3mW7Rs69MzEI5vVvD7rign7em9WKv68F32wZzehLWtKq0rUhzbrBHFweNSC5yrP1a0gqrD+LfLn8P4vcUlhP+AuVje0fevZM/FiKasyaxOYU1md3ro99Wr/LZyaQe1+/FcRu6E9hSnv/0OxXt/Yi1i5DI+DpensScn+cq86212fcMaCBy/7xWibSXEsQen+RW87l2zeX13rGQ96N7q/NnStjmdbmX9ac/Xq0LrXs+/We/bxPt8ZBM+B9L3sB5U895LKN7+IJ8Ts5axp+zkxqxt/WM9b6/PbvbVdHq0Sc7jeXewdnriP3ndub7gd7NQt+4F1kqPWfJxdDWY+66IngYzfGz56kVmGIZxMAleXIyDg/UiMwzDKI2U415krZ1z32b3IfveOdeheIdpGIZRzigDRsuIGoxzriOAnQBe2+91cc59gqxeZB9lN7q8w3t/RqSNBeeDUZ9LrDxXfTIujq+F6nP5ejHnT9Un42pxHlm9LlP+FxqBemTGP8r59ALP57KCt6U+l8Pe41yw+mR28+qJ6jedQbH6ZN5p8wDF6nOZ9TZrVeqT0V8gul8zlq2mOOiTyczgHaU+l+0jx1GsPpnxFfmsUJ+LzwifVv7Hr6wtqNclMaBVqEemYjPuZac+F79gDsXqk6nZlXubqc/ljQqsY6hPZtmC0L5Qn4t6RdQncwPv1lw+l0g+mSYnyr4I+FwA4IsnWD8K+mQqyLwaozIkAAAgAElEQVQ36nMZtpKPcSSfjPpcTm/Geqb6ZOZdyfqoel2++L+VOY/VIzMsiXVZ9bkgUYQ6+awkDXkyuhrM8L7R02Due7N0+mDy6EXmAez/pqkGYB0MwzBKMcGLyyFBGagiK6wGcyuAj51zjyHrR+4p0RuSYRiGURaMloWtIrsOwN+89w0B/A1hfDHBXmQfp67I62mGYRhGGaNQvcicc9sAVPfee+ecA7DNe181zCoAAOMDvcjU59JZ5hVRn0zH47hRZi6fSyOejUF9Ml0H8+2mel38plCeWj0yW2M5fak+l77/4Tyv+mRikuT14nNJW8Qii/pk1nRgTYaeO+FZitUnk/7FtxSrz6WCHAf1yZxxN2sJul87HMXaQdAnE9ezBy1Tn8vK1TwW9cko6nP5x5G/h33+T8tZg1Gvy0NNQ9tTj0yL17tRrD6XLd+yDqE+mcWP8pws6nPZM491EfXJ3NUktF/V53LCAD5mkXwy6nO5KZ11EfXJuOo830vQ5wIAO9awLyzok2l7HJ8P6nPJkGmK1Sdz7l7WptTnEsknU3XgyRSr1+WFiqF9ox6Zjnfy+1afS/2bm1OsKayk656Oqs6xc2ivqN3CVH7o7UPKB7MOQKfsx50B/BTmuYZhGCVO8OJySFAGqsgK24vsGgBPOufiAOwBMLg4B2kYhmEcehS2FxkAtI3yWAzDMIz9lAGR/6D2IpsU8MGon2LnZJ47RDWWqeM4P6o+mQ9Fs1GN57S20qtMvCqv/T2U5VONJUkOtPpk/GYeS6ReZgWd06Xh7JAmox4Zn85jU31n2wes72gvsc/e4Ry4HpepormoRrMjhnPsQZ/NrkW7EY5I87moL0Y1mrRU3rb6bP64awLFqpNkrAjtG/W57HzwdYpVoxkVx7/N1GNzbHPxnohXZde3vFx1lokVQ+tXj8zKRaxdqb6z6aPwvcRO6iBz04jPJeUZPp9Vo7lhPKfzyWcTYT6XD676imLVWEas5/emPpmEKtxnTTWeGlcdT3FQJ1n3NGub6nO54Sn2iKlGU8XxttVjk3jlQ9HVYG7vET0N5rH3DikNxjAMwzDCYr3IDMMwSiNlIEWWn15kDZ1z051zS5xzi51zt2T/f+/sONM51674h2oYhlF+8Jk+an8lRX56kdUDUM97P885VwXAXAAXIatdTCaA5wHc7r3/PtLGUp8ckrMx9VOcdTl7T7TXWGIL1mDUJ6N+Du111v0J1i7UqzJ5fmhOF+011v927kulPpnujzamWPucxV83guKCzumSGJhuRj0y20Z0oVj7nMUdlkhxsFcYAFTv04xiPS7xorlor7KuPTlHHvTZdGrKupcSaT6XSJpMuzFnUaw+m5qXcY5c+4Ud9miofkV9LssmsW6mvcqW/shjUY/NE5eF96qsWMreFO03FhvYvHpk7r+Me2hpn7M/0tlLEuwVBgCH9WHPmPpcEo4L36usYVPeF0GfzajRfD6qTyV9L58/2muswxF8zNUnc2891km011mVS1tRTP3CUlNpkfpcVvzJ29JeZc/ex98h6rE5dcPEqOocO269MGpXhipPvF9qe5GtB7A++/EO51wKgCO899MAwLlDbJ5rwzCMQ4HykCILku3oPxHAdwV4TU6rmJe+XlKw0RmGYZRXMjOj91dC5PsC45yrDOBtALd677dHev5+vPejvfftvPftBp5yXGHGaBiGYRyC5LcXWQUAUwB87L3/jyybgXxqMBu7dMrZmPpcPn2DvSXqx3j/4R0Uq88lTVJ16pN54x3uqaVel+6tQ56IXPO5LGaPjvpcJv/fKorVJ7NtIusmSqQ5XXbefF3OY/W5VLv3M4rVJzNjOue4db9uHb+MYvXJaO8ynU9mmvhogj6ZDxN5Jxd0PhfN16vP5ft+n1KsPpmkFuHndPnj/0ITp6hH5uYnWLdQn0vz43lb6pNRnUJ9LsdE8MkEdRX1udwzjv0Z6pNxLB/l8rn8MZ5n11CfjGoPOp/Mmp/4vQZ9MlNfY81PfS5xCfxrWn0y2x8aT7F6nf75e3ifzL5UzvqT1yUpiZZpL7Ef7uWxqs/l+uH8OVafzLm/R3fOlR3Xnxc9DebZj0qnBpPdzPJFACl6cTEMwzCKiTKgweTHB3MqgP4AfnTOzc/+v7sBJAB4GsBhAD5wzs333p9TPMM0DMMwDjXyU0X2JYC8bq/eje5wDMMwDAA4mG28iouD2ots523dczamPpfK3ZMpVj+GXuHU53K+aC7qk9kRw89Xr8uV/w7lW9Ujk3TZaRSrz8XV5G2pT6bj0eFnlI40p0vmitD21Ofi4viNqE9m+6MXUqz7tYv4WNQnE0mTqZLJPoSgT+bUZPbBFHQ+l2rD+1KsPpcKSbztXD6ZcedSrF6X2GNCBiP1yKz+jXUG9bkMSee+VOqTWS4+GfW5VPoL+2DUJxPUZNTn0qQl62Dqk3ngWvbBqM/lu9n8fPXJJNbYR7HOJ/NMH/7OCPpk6jXk80t9Lhe8eirF6pNJ2816kx7jvTv4N7H6ZJ66meOg16X+TTKvkni6nGg06nPZ4XnbqlW9vCq6c65sv+bsqH05V/3fJ9aLzDAMwyg7FLpVTGD5351z3jlXO691GIZhGAWkPEw4BiAdwN+DrWKcc9O890uccw0BnA3g1/CrMAzDMApCSfYQixYF1mCcc+8B+K/3fppzbiKA+wG8B6Cd935TuNe+F5gPJrkG55G3bOP8Z6tHWJNJHf8lj0PuvXas4Zr0Oje0pjj9R/Z76FV9ysSQjtLtEtZzpoqe06oaj33pVs77dr2HdYoZ9/Fu0X5e6uFRr0lQFynofC5V/+99itUnUyGZ+1LFNGpIMSqyryGuS3+KX2k9HHlxdh32vRw2hOfq2DVhHsWZLGtgWUodituPbETxF3f/RvFxDVnneHgT+ztGNGHNJ2VJaP1te0kvvCU8l03lM1hL2DmDPRPLlrLmcmJf1rJ2zOb1P7yO31v/NNY93qkQ2u831edtVWnFGkvcxd0o/vxKnnPlpNasAWbs5fNt+dLwyYc2f+Xn75jF+7lq75CBetBjPNYr9/BY28tcNEm38fmEjdK/bif734bcv5LikYez9jV0A/vdHm0U+ixp77sM8VnFqkfnH1dQ/M3l7LtqfjTvh/pfT4+qzrFtwFlRu8JUe/nT0umDCRJsFeOc6wFgrfd+gfUjMwzDiDJl4A6mUK1ikJU2uxtA3j9fQ6/L6UX2ceqKSE83DMMwgKxe9dH6KyEK1SrGOXc8gM8A7O9/3QDAOgAdvPcb8lgNliefm7OxlC2cRjr7sSYUL7yD27M0bc+llNLlAUkXtad44zPzKa55mtSZStly8NdCMF0GAG2qcspq4TYe+wX/4DTVtPv5+VxomXva4dN7yfTPEdqzBFNmkdrtn3EmpyS0jHnH/zhFkbmap1iOv/5+itM/46mEx/ydSzmVK4aHUkd/jPqRllWsxmWeMXI/XbkvH9M5w1ZT3LgB75clayRNdRy/93tX8hTN/zo/lH6Z+za392l/HQ8mUkqsWXNOl2RIHXyVDrz+pRN4/a/LtNp31d9I8dPrQim64c+fQsvS351C8d6VnN77bj6nQas63u86vXPlU/m9zXuRvyPaDuP03vYJ3MR2/rJQOfprFffSshdfuYDi1P/w+aSfY1TmaTr+/M8XFA/7nT97/zqSvyf+bzV/lv51RCilFikltv2BsRRrSm3pz7yfOv/+VnRTZP27RC9F9vpnpTNFdqBWMd77HwHUCTxnFfKhwRhFR30qpZngxcUoPMGLS2kneHEpbQQvLocCZUHkL3SrGO/9h8U3LMMwjHJOebjARGgVs/85jaM1IMMwDKNscFBbxXxXv2fOxrQM+ZPbufxQy5gTErmGVcuQUyfNoVjLmD+e04BiLUUmTUZ+OXz+Fpf+ahnyBw/wrbeWMc/fFl5TmfU2az5axqyVGMFS5Ejt9mPq8ra1DLnKNZwD1zLmCbOOQDj6/ZtbmgfLmLdeNoCWaRnynm1cWq5lzDPvYd1Dy5BX/cbvTcuYv7ubNRstRb7zw1B+X0uY4fk3VSTNRcuYh77C+X0tQ27em3eGljEHS5G1DPm+a7+mWMuYY2UuCi1D3u55v2sZc/pufu/BMmQAmDuS9aFgGfOehbxMy5D/evUHFGsZ8zhpv6JlyLVu60SxljFvH8PfA6qb3Lk2pNkES5gBYMky1pa0DDmSZpPYZXBUdY6tfc6M2pdz9fHRLaHOLwUqUzZKHvW5GIaRP4IXl0OBsqDBWC8ywzAMo1jITxVZQwCvATgcgAcw2nv/pHNuPID9NbLVAWz13rfOYzWGYRhGQShB/0q0KEovsj77n+Cc+zeAbXmuIZug7qI+l+QaqRTn8snczrX86nOpIh1O1Cejmot6XWi5eGRUc1GfS6vqnCdWn4xqLupzqeD4Vlh9MmfcHWp/kbvdPmsuWsZc41reMepzUc1FfTLP1QmvwWjrmKBPRjUX9bmo5qI+meMa8n5Rn0unB1lXU5+Mai7qdRlxXEh3UY9M3z287Uiai/pk+qdxOiaXz2X2VorVJxN3XmjuPvW53FSfvSVaxnz+HtZ71OcSm8DvTX0yZ4qvS30ubf6at0+m3Tg+H9TncuUeHqv6ZB45nD+n6nMZJZqL+mTCtX957Kg/aZruXB4Z0VzU53LyG2dRrD6ZxC6DEU3KQoosP1Vk6wGsz368wzmXAuAIAEuAHJ/MpQA657kSwzCMEiZ4cTEODoXuRRb479MB/O69D2/pNgzDMPJPGUiRFaoXmfc+mKe5DMC4MK/L6UX24uc/FH6khmEY5QifGb2/kqJQvcgC/x8HYC2Att773/J6/X429+iUszF90zvXhm+3v+axRRSrRlM3hrUH9dn8+gC3hledJD6wH1RzyVi2igcrGs0U8cnk8tikczcy1Ulm/8J5aS1FnhrQXVSf2RETS7HqPXGd/sJxAdrtA8B1G6dTrJpNwrknURzUeOY+y1qA+lT+eI5/cKjGknwEaweq2ewYx8dUNZ7MND5OqpvMHRUSiVRjuXcVeyLUJ5OUzNNLqL5z7NE8dtVJ7j5XjpN4Xbbc+3bOY9Vnhk3ibavHZk8GnxPabt/V5f2sGstcmYJZfTIpC/j1QY2nav92tEw1kiF3LqRYfS6LZL+rRvN0K95v6rPZfh9rPupV2XJP6LewaiwvV2TRUH0ymem8H9UXc/j0L6Kag/vzgk5RE2FqfRDdseWX/MxomasXWYCzACzNz8XFMAyjJAleXIyDQ1F7kfVFmPSYYRiGUThKMrUVLYrUi8x7f3W0B2QYhmGgTIj8B7VVTPCKrL3CVHNRn0vKFs7Naq+yhsPaUqw+m1896yraL6zhTUfnPFafS+e+ec8dA0T22HRqzDls7Rd2dh/WXNTrEtRdtE9ZJI9Nz/vy9qnkh0g+mW1hplxuP5Jz2upTSfWcA9deY7UHt6JYfTJJPCNyLt9NJK9Ks+ahx9prbEQyay7qk7l0icyRIp6bBz/kbWu/sLiLe1GsXpeg7qJ9yvqnhZ9L5qGr+ZtJ53NpeixrLtpr7KQpiylWn0y7ZH4vwV5n7URz0V5hqrmozyVNNBftVaaai/psws3pEpsALF4SOo7aa+zRON5P6pN57Ciea0Y9N0ZurBeZYRjlguDF5VCgXKTIDMMwjINPWbjA5KeKrKJzbrZzboFzbrFz7t7s/2/inPvOObfCOTfeORcfaV2GYRhG+SGiDya7TLmS935nth/mSwC3ALgNwDve+zedc6MALPDePxduXavbnJWzMdVcdk/meRz06r1iLifc1eeyZuRcitUno6jXZc3TP+c8Vo+MlL/n9rko4pP57E3OS6vPJWMZaxPqk5m1MpQDV81l2jucw1afzNoK4W9Sw83nAgDpH4ymWHuZVXtgBsVBzWb7T/z7RX0uSY79QeqT0bk91OeS+mf4+WS+H87V8+p1CaJ6TeZG7hWmPpe3EvkY5/LJNEukOJzPBcjtk9mF0L5Rn8sRjcL3MYs9mvvPqc/lp+U8P4z6ZFSz0V5m36fk7ZP5+zJed675XAa3oVh9MqljPqdYNZcv+n5KsfpkHjuC903Q69LiOD5G4TwyQG6fy+2/8HeQ+mTqzpwRVa/J72dGzwcTbY9Ofol4B+Oz2K8yVsj+88jqPTYx+/9fBXBRsYzQMAyjPOJd9P5KiHyVQTjnYrM9MBsBTAPwM7La8++v2/kNWQ0wD/TanFYxb2xaG40xG4ZhGFEm+3v+B+fclAMsu9o594dzbn7236D8rDNfIr/3PgNAa+dcdQDvAmge4SXB144GMBrgFJlhGIaRNyUg8t8CIAVAXtPmjvfe31iQFRaoisx7v9U5Nx3AyQCqO+fisu9iGiCrJ1lYgrqL+lwqy/2P+mRUc1Gfy4ZM1k3UJ9PgxmMoVq/LPhd6vXpk6p/NvZ1yzSVzKftW1Cejmov6XDocxZqL+mTObx0aj/pcqnjWMdQnE4lw87kAuTWXoM8FCO+TmVLjdFqmPpfDrjuRYvXJJLOkksvnEmk+mWbN+QXqdWk7JHT6q0cmJUWeKz6X5BSev0h9Mv++qyPF6nNRzUV9MmtXh+YAUp/LSNFc1CcTMy+8z6XNBj4OkTSXoM8FyN2bLOiTGdmYdY5c87lE8MlE8rm8VpHPf/XJaH+xoNclnEcGyPLJBFGfi2ou6pMpmMMsMj7z4KW2nHMNAFwAYCSy9PWokJ8qssOy71zgnEsE0BVZV7npAC7JftpVAN6L1qAMwzCMg8oTAO5A+P4BvZxzC51zE7NnOo5Ifn7u1gMw3Tm3EMAcANO891MA3AngNufcCgC1kNUQ0zAMw4gC0WzXH9TCs/9ypt90znUDsNF7PzfMcN4H0Nh73wpZOvyr+XkP+elFthBZk4zp//8CoEN+NmIYhmEUDB/F6q+gFn4ATgXQ3Tl3PoCKAKo658Z47/sFXh/sk/MCgEfys92D6uT/8u5QPn8fOFd6WouNFMf3vZDifW++T3Gdw/jm62kuf8dznTiefh/3EfowkfPz9x4ZytVWu7cfLcPPnNPu0ZJz1jc+yLX4DzXmHDZ+ZL2olrz3TJnKNa5nD4q33fdmzuNOTVnqeuBXzv3f04SX79kW3iuy9bIBPBbROZalcA847S+mXpeg7tJtyyxatqkz6zWowtpU25vYOxJzDM/CHdfxUor/12Ekrw91KTpzB+fMW1/AJ8lPL4e211CO6fcJnJBvvYWPaXxtPmYdlrNOkvrUBIpf/pEzCneeyscpoe95FO8e9lnO4/vbsmb3zLus33SL5bHPSKtOcf/vl/HYzz6F4pan83FaPZuPS6uu7GV5azq/l5MSQ76wSg04w/LUsbzPd4xmn0uVy7iHYObnn1CczvISOmVUovj0G/i9v/o//ko76cIjQ8FG3ueVO3GPtb0/sLZU9fIzKH75dp6097QyUrLkvR8KYCgAOOfOAHB78OKS/f/1vPf7hcLuyJJJImKtYgzDMEohJd0qxjl3H4DvvfeTAdzsnOsOIB3AZgBX52cddoExDMMohRzMKrKcbXo/A8CM7MfDA/+fc5dTEIrSi+zF7P/bX1VQOdK6DMMwjPJDfu5g9gLoHOxF5pz7CMDfvPfbAcA59x8ANwJ4ONyK9gWuZ/FSDRdJc9m8kK+FNVvx65+rxL6E6z7iecsriebSfTe/9aDusm3EGFpWtR/XODjpNaaay9BV7BsYwDYXHJvMPoSqwy6j+M9hb1EcJ/X5Qe5psoHi+1eyDvHI3ZxnVq9IRbb05Or3pf3BCjKni2ouueaSacm5/Jim7FWKO703xekzeb9E4oiOaRSvncl6VMOWIe1gzSLeEYNEI3nhKzZq1Wb7EXqcwn3PXvyO39uA49lPlND3Aor3vvkRr79t6HzfNJcPSrdY1jWmZLDmcl1n9tS8/jlrNleksuZS8SL2KzWp+DXFK2ey7nFSHPfi+y49tO8aJfHnMDOVd5RqLjvGceGSpoUqn8YaY78zWfgYM5o/i1fdwJ61V58Nbf+qv/L5F9OMz8/EZPaP736D9aJYz8c0o5hvMCK0iTwkyE8VmQeQqxdZ4OLiACQiqz+ZYRiGEQVKIkUWbQrVi8x7/132/78MYAOyWsc8ncdrc+qvp+5eEaVhG4ZhGKWdfF1gvPcZ3vvWyGoJ08E51zL7/wcAqI+skrU+ebx2tPe+nfe+3bmJxxzoKYZhGIbgM13U/kqKiPPB5HqBc8MBpHrvHwv8X0cAd3jvu+X9SmBb/y45G1PN5cuB31C8T65942Teh+fOk1zvLs63q2aTWI2Xq9fl874h38Fk0WvSJPunmkvVfjzHhZdeZJtfXEBxrZHs59g+kuehWC59sMZXDOWVVXPxERLBsQk8FvW5RPLJbBq9kOICzekiPpfMn/gOttq9n1Gsfc3e+PKADbpzGPCYzGUjPpnFf7mT4qDmAgAbl4e0BdVrMrZz/N7XDXisnOrPpdnonEAJl55LceroDyhWn0xwDiLVXBKTeGy12/JBVZ/LnkmsuYz9lt9Lf9Fs3vpM5nsRzSVTDIBNOu7Kebx3HY9FNZc9H82jOCaJd+Sb03nb/QbL91Mav/edX8ocL4PP4LEuC+kur77I27rq+rz1GgAQexF6t2UdLfFy9mkl9hkR1W/ylSd0jZrs0GTBtNI5H0wevciWOeeOyf4/hyzjzdLiHKhhGIZxaJGfKrJ6AF51zsUi64L0FoAPAMxyzlUF4AAsAHBdsY3SMAyjnFEWRP5C9yJDVv8awzAMoxiIZi+ykuKgOvmDuov6XPaBe16pT0Y1F/W53F+d55VQn0z8QNZc1OsyOTGkLahHpl0y56jV5/Jf0VzUJ6Oai/pcVq5mXUN9MvcEHqvP5R9Hcg5aUc1FfS6R5lRJWZu3zwUIP6dLrt5i4nPZ1Itz2uqTeerw8BqMai7qk1HNRb0uR521O+exemS+S2X/hfpcXBwfY/XJ3PB0K4r3vjWVYtVc1CezPiU0b0oun0tbPh/VJ1Mzgs+lfxL7XNQn0zGMzwUALj2Ttx/0yTS7h4+x+lwS6vNY1Sejmov6XC5qyue7+mSCmgvAXperrl9FyyJpLipvquaiPpnEPiNgMNYqxjAMoxRS0r3IokFRWsU459xI59xy51yKc+7m4h+uYRhG+SDTu6j9lRRFaRWTDKAhgObe+0znXJ2wawGw4e5QaWZNziBELEOe+S5PvXqVpNBqtOTb3fi+3Sn2Py2muNqVXFqc9sCvOY81JTZ8FaeJekkKbfuYH3jd94ZPx2nrl2AZMsApMYDLlvsgg1Joaan8Wk3HzenHpcDa+uWLuzn1c1xDzhMkH8El2bUH84HbPmYOPz+QccvVbl9av7z89/ApsZt/n06xljHve+khijUFp2koLSV+Z2poe5oCO4ln9EZsVU6hpW3YR/FZjlvq73vnY4r/XMCvT5fPfDAlBgBT00NpsThwCs3v49+F9Ud2oXjZzV9SrK1fVs/i9LJUsmNmBqfE+nfhz4OmtZo9eELO4+0vzKRlmsIa8zJPa6ApsfSfuAz/0pN4W2kyLUdMIy653v0+p+SC7V9ijjoKr/090HJfjoGmxLSMOTOFC2UT+54BIzyFbhWDrKqxy73PupHz3m888BqMaKL6jFH2UX3GKBx0cTkEKAsif1FaxRwNoE92G5iPnHNNw6/FMAzDyC9lwclflFYxCQD2eO/bAfgfgJcO9NpgL7Jxf/52oKcYhmEYZZBCt4oBMAjAed77ldlu/q3e+2rhXhtsFaOtXGqfzPlOLUO+ag/nsLWVTOeXTqJ47xuTKU44szXF2s5lxxshHUXLkHvt5m1rK5knXj6HYtVcIuGkPlJLkYPtYbSNTLsxZ1GsJdC1BrPWpO32GzfgaYW1FUzH+7mEVcuYk2px645gWXTNN0bRMi0j1ilolctPY81Ey5h3jLmWty2taNIWcWpJS4n/elLoB4+2gul5LpfDahnz3n2cXdaS6K2rK/LYpZ3Lmllcwq2lyNe0DI1dtaRcJdDS6t/F8WdD2+3PyeAWPtoKplEn1j+1jPmvDzehONiKJr45S7HayqXymbyftQz50pP4R6i2konvdALFWgZdZXD4UuI354X2pZYl9wvT6h8ABjxyNMVaEl3pn+OiequQ0vT8qLWKSf7pw0OqVcxSAJMAnJn9tE4Alh94DYZhGCVP8OJyKFAWUmSFahXjvZ/inPsSwFjn3N+QVQQwqBjHaRiGYRxiFLpVjPd+K4ALcr/CMAzDKCol6V+JFgXWYIpCv0Y9czamPpdN33C+U1u9KNru//OB3/Fy8cmMS+T1a8v9KpeHrqHa6mXba9xiXH0utw5gz4O2mvkokcdS0Jb76XtDmUz1uXzf71OKtYz5hyWcP9dWL6t+q0mx+mTU56KtZlL/zLvd/yv3hy+pjtRuP5LPpUq/5ylWn8ykmay55J7WOKQHqEcmYzsfM20zkxDPmoq2+3/2U/Z/aMv9hqfvpljbvdQ6IbQ+bfX/zE08hYKWMa9dzBqLtnppH8uenWC7fQDI3MOfFW018+JdKykOtvt/87Pw7fZ3TudjoD6ZXV+zZqPt/vd9wVNfqCfnre9Y49GPVt82oX2lrV/U56JTKr98x88Uq0+m0tBXo3pF+LHJhVH7cj5+5fulU4MxDMMoCwQvLsbBwXqRGYZhlEIOYnKp2ChKL7LOzrl5zrlFzrlXnXN2sTIMw4gSZaEXWUQNJtvjUinYiwzA3wCMB9DFe7/cOXcfgNXe+xfDrWvHkHNzNqY+l2c6ci2++mTqPsj1BNruf/Yn7F1Rn0yklvvP/OPInMfqkXFNW1CsPpe5yzjvrD6Z+5pwF52CttyvNrxvzmP1uVRI4hy0+mQaNdhCsfpcOj3IOWv1ySQfz2PX9v9VLmOfTdAn88Ef/D6Va2YPo1h9MnD8wYjkc1GfzB89WONRr8tFHUO6i3pktLeY+lwq1OWeWuqTycjg8099Lr2rsD6Va9rjniFvlYBGGoAAACAASURBVLb6z9jKx1x9MtVEvszlczmd9U/1yQR7iwG5p1xWb0rQJzPgmvDt9vsN4B5uuaY8voI1F/W5VKjO69OxjJ3Bn0X1ulz579A5kavdvvQWU58LEviYq0/m+jVjovpNPr9R96jdw7RePbl0ajA+C+1FlgFgn/d+/xGYBqBX8QzRMAyj6AQvLocC3ruo/ZUUhepFBmA2gDjnXLvsp1yCrM7KhmEYRhTwPnp/JUWhepEBaAGgL4DHnXOzAexA1l1NLoK9yF5eYlUchmEY5YVC9yLz3j8W+L+zAQzy3l+a9yuB/gEfzIhKXHvv5FJXb1BjijOWsjYQ2+Ioimc+wD212rVhzSWGW0Mh4YT6FH/xXGg/tG/Fr/3HT9xf6e6arGts2MC+g5aDecKXlBfY89AomV8/9CfWRa7aw2MN6ii1+7EXJGMFX7TjevWk+O8DPqF4RBPOeacs4ffWttdOilOX8Ngrn8E57rmjWDto1jykLWxeyzqbekV++ZT7canOEWk+l0g+l8Pe415n6pPZuyEkVlS9QzK8FTjf7hJ5vpbUR0YjHPHHsr8IGfz76/mJfM7Ukp9nWwLSwuD+PFeS381xhSF3UPz0WdwDrp7oZt0v0UlVJIUiGmT8jXdRPKYzr7/vwNAGRr3CtT4D2/MxUZ/K5Rfx53bjDN4RNZryh2H1XNayGp3I7+XtuXzOdK7OWteU7aHP2ikZ/B10VGseS9KArhTvfZvnVkq4+EyKEy8dHtVc1PcNLoravUe73yaVTg0mr15k+ycYc84lALgTwKi812IYhlGyBC8uhwJlQYMpSi+yR51z3bL/7znv/edh12IYhmGUK4rSi+z/APxfcQzKMAyjvFMWepEdVHNkUHe5dxfX3j//N759Xf/CKorrXctzMWQs/oXidm1YK/h+HmsFp1/HB2vvgnUUtw9MsTFnIb/2bpmX/sHNNSj+z2DejYtGc478qBasuaxO4ddfJTnyV0Uveiqgu2waw16Qwx69jOL0t9+heEQTzjPfu5J7P/2r1w6K575dmeL213HOe+cM1qeaNacQy5aGjmPrCzg/rl4R1Vy035dqLrnmcxHNRX0um3rxMc81n8x7d+Y83v7I27Ss2gODKfa7eT9FRDQXxLJfQzWXP3kxhgR0l9Gvs6Z3/ae3UJw26hGK66WzvrNeP+UF1Fz2/fdhivsO5HPkzZdCGxgo87m8NIePyWDRXN6YxFpV7/asKW75iT8Mqrms/oHPT9VcPt8a0Fw8fxa+juXvoJYD2IOT+vI0ipOu6kLx3nenU5x46XBEkzJg5LdWMYZhGKWRsnAHY80uDcMwjGIh33cw2SL/9wDWeu+7OefGAmgHIA1Zxstrvfdp4dZhGIZh5I+SrP6KFvn2wTjnbkPWBaVq9gXmfAD7JwN/A8BM7/1z4dbxc8tzcjamPpdrH+fcqfpkVvzJuoX6XBIasadCfTIzxCejXpfYQFpZPTKqqajP5bZXWEQZWp3zxA9sZW1BfS6Jcfx69clUbBbKFavP5Y//G0ex+mRmP8TvW30ud37I/g71ySxLYZ9M0OdyIII+mZ9e5mOomsvG5ZwDV5/MO1NZLwo3nwuQW7MJ+lyA3F6XKj3+lfNYPTKZqeHnI0q8ibUv9ck8c9lHFKvmcsnlvG9yaTYB1OfyrPhc1CeTuYP1SNVcnn+Hz0f1yeyRvEbQ5wIAmVv5HApqNq+IR0Z9LqMns+aiPplVc3hsqrls+Zk1GfXJvC4+mKDXZb7j8031mlrHc5809bmkvso+GPXJJF7yj6heEWbVvSRqMszpGyaWTh8MADjnGiBr9soX9v+f9/7D7D5lHll3MA3yer1hGIZR/sivBvMEgDsA5PpZl91huT+Aqbose3lOq5g3N/92oKcYhmEYgoeL2l9JkR8nfzcAG733c/N4yrPISo/NOtBC7/1o73077327vjXtJscwDCM/ZPro/ZUU+ZkP5iFk3aGkA6gIoCqAd7z3/ZxzI5BlwuzpvQ+ftAaQ+sS1ORtTn4u+Wn0yjx/3J8Xqc+n4D87tqk9mz8+cl1avS6eAT0Y9MpE0mcPrsEfioa1cm99vD+9j9bk81JRzweqTafF6t5zH6nOJPYZzzuqTqdGBe2qpzyX5OJ7vRX0yIxrz8qDPBQDaDuE6kaBPJq4Sv2/1uRx1Fh8T9cnUbc3zloSbzwXI7ZO54fkOFKvXJaFu6PeVemQ2nsdalpJ0R3ifzPirvqRYfS7XXsLzzahPpsI1f895rD4Xl8gaoPpkBvVkrUu/YSZP5PNTfTLXXc2aS9DnAgD9Ph9CMflkMvmDrD6XSyP4ZAacyD4Y9bk0aiuajPhkNm/h/ndBr0u3qnl7ZACg33+kV534XBJ6sQ9GfTK1PvgiqrcKMw7vHbVLwxm/TyidGoz3fqj3voH3vjGyOih/nn1xGQTgHACX5efiYhiGYeSfTLio/ZUURfHBjAJwOIBvnHPzs7ssG4ZhGFGgLGgwBXLye+9nAJiR/di6ABiGYRh5cnAvEoE8c53T+Kr61buca31cfC5bV3OudWxFrllv+vzPFKvPZte3rMm8l8i+g1qjQ3HLway5KE1PYj3o269Zz3lAPDaVLuCGXY1EJwn6XACgxd3dKN754Os5j1WfGRXHY7lqDy9PqMx91HL3FuPj0Ff0ontXsQ9mRDL7ZDI3cs47JSWU1/4+QbQB8alkbOdtfZfK+s9J8ylEz3N522nicznLia4hc7pof7HgnC6qudT5iI+R+mT8dt7vqu/0vpr3m+okSlBzAUR3EX1mlKxLPTaReotdVP2/FKvG8pzM6VJPEuD7nv0X/0fAZ5O2hv09Z1XnwanPRTUXF8fn4+dgf1HlORyrz6b6So6pv1gmH8PeorGMuY1CdK7O58/mmVzn9HUs65/cIa7olAXdwe5CDMMwSiElmdqKFvnWYJxzsc65H5xzU7LjV5xzK7P1l/nOudbFN0zDMAzjUKMgdzC3AEhBVpnyfv7Pez8xukMyDMMwyk2KLNAqZiSA2yI8PU+C3hTtFdZu9WKK1eeimov2KlPNRX02j+xk7UH7hR12Ue2cx+pzUc1FfTHtd7JvRj02LbdwPl/7hbkT2lOsXpeg7qJ9yq76kWv51WPz5Bk8Fp3PRX0t2mtshBNNRnwyly7h5cFeZ623sP6jPpXaoh1or7HYquyLUZ/M3n3cf057nWl/sILM6aKai/pk1m5gzUX7nI26djbF2i8s7iruL6ZeF9JdpE9ZpLlkIs3n8tY7/FnQXmOTXuCvhVzzyYRBfSnaK6xGJsfqc1HNRXuVxdVjzU99Nlc+cQrFQa9KbFVO2Givsc4/fsVjEZ9MquznYJ+z4qAsXGCK2ipmpHNuoXPucedceBXTMAzDKFcUpVXMUADNAbQHUBPAnfra7Nfn9CJ7ecmaAz3FMAzDEMqLD+ZUAN2z2/NXBFDVOTfGe98ve/le59zLAG4/0Iu996MBjAaAqYf39V8ty17w7mZquZ90fgveaCMuK276JadiNCW2bvQqirXVzI1pnFcIpsQAYNPkUGqobl2eFrmHlCFrSizpVE6Z/aUyL69wBI9FU2J+wRyKt3zL6ZRRcYG0wE+H5Wr33/z40NiHSBnzzhmcpoqUEqssKbXMjZxK1JTYW4l88iancHuX+Nqh5X896Tdq97JJUg5aopq2gdOimhJLiOfUToW6XFYaLEM+EOFa7msZsqbEjvhKyphfe5fiay7hVI+mxNJffZLi/03kaY41VZCr3X+fQEm2lDGnj+EyZE2JXdqTj2lcP06pdd8uZchS9vy8rE/b/R8XF0pFbp9bkVvuy0/aqlX5ZK68VfabpMTS1/P5tTOGU2TaqkbTYhnbQ8tTX/2MWu5rmxlNiSXJQalZg8cSbTIP/SKyIrWKqQcAzjkH4CIAiwqyYZ3PpTQRvLiUNsJdXEobwYsLkLuXWGlC9ZrSRNiLSykjeHEBcs/nUpIELy5A7vlcjOhTFB/MWOfcYQAcgPkAhkR4vmEYhpFPSrKHWLQoSquYzsUwHsMwDANACXbZjxoH1ckfTItpGbJqLlrGXK855wm0DFk1l1xlzJdyrlZLkV+KD6XF7q7JpcBJtXgqXy1DVs1Fy5hjOnGbby1DVs1Fy5ivepj1gGApsmouWsac8mP4KY9Vc9Ey5mDrFyD3lMuquQTLmDssZ01Ey5BVc9EyZm39omXIqrloGfPhx7GGo5DmIiXM2vpFy5BVc9Ey5l2LxlOsZciquWgZ84RXuCiTSpFFc9EyZm39omXIqrloGXMuZMpl1VyCZcznSUpMy5BVc9Ey5suP4FYvWoasmouWMe99l9sRaSkyYkKqgLbb19YvWoasmouWMR8NQ7FWMYcY6nMxyj7qczEKSUxRmscffMqCD8YuMIZhGKWQTHfoazBF6UXWxTk3L7sP2ZfOufBTABqGYRjliohTJuc80bnbALQDUNV73805txxAD+99inPuegAdvPdXh1vH8uRzczZW70y+Om+dw/nyOWu4JckJtVmHqDfwSF655KE3jmVT50iZxvj+I1mLqHZvv5zHmZ9+TMt2z+NpgxNP5VztV09zUvptmQogCZzjGNaMdY6KbdiTk/Ia3xwf2TSkq1Q6uRYt2/Ut+1xWiGbSqi/n9nfMZg3l4XWs0fRP4+NwRCPOqT+9jjWbu8/lnHncxaGpBlKfmkDLXv6R99ug00W7uvRcive9w8dh01y+4Z6wg99rt1gea+NerMnoOfJ8QAfRUuDeV/N+03b718iUxxWGsM+lUss+FGvrmcT7RlKc9tK/KY4965ycxy6e/T8vXM77Rfnrc20o/vP+KRR/sJY1wn438PmZNv9XijP38fkYV4fztDFVQvHaKXz+HH4iaywrZnG7/h/AU3hvl5+8g7psoDi++1kU73ljKo9Vppt4eQGfc3UCx/mibvwdkLmdx75xPh/zbdv5OBzdgc/9mu9Gd8rkCfWuiJrO33v92NI5ZTJAvcheCPy3R6jxZTUA6/R1hmEYpYU6Om9OKSczin8lRX41mP29yIJutEEAPnTO7QawHcBfojw2wzAM4xCmKL3I/gbgfO99AwAvA/hPHq/P6UU2fqv1IjMMw8gPmS56fyVFYXuRfQCguff+u+znjAcw9UAvDvYi23H9eTk5xfXTOb24aAtrLu0b8vS4tS5jX8z6lzhPHBvHN4J1ruDc6/3vs8/mnl85f/9EQHeJCeS/ASAxcQbFu7/iC+Xbiaxj9NrNOe02rVlzGbmMdYzLF3G+P/lKve6HdJdd37AWtUJ6ix2TzHnlHbN5XVU6cM67/wTOO78u0wxXEs3lpvr8XuIuZn9I+ruhfL9qLgOO5/2WcOkFFO99i0+hPxewhlK7LWtd3Wax5jIlg3W2GzOkpYr4R4K6i5YCq+ZS0Hb7kdr9b23MmkuFgTxlst8QmgLc79uNgqCaS617eArufj8uoHjMM5w76t2eP0sx8XwOBTUXAMjcEdJZDj+Rz6fff+DnquZyIlgTbNqF4xc+q0vxwJ18jlS8nHW7Z27hjlUDTuBzrkL90HgmTeHPzl+qsp5Zp7X40/bwcfh5NntypCtakSkLTv5C9SID0ANANefc/k9RV2RNRmYYhlEqCV5cjINDoXww3vt059w1AN52zmUC2AJgYFRHZhiGUY4pd61ipBfZuwDeDfd8wzAMo3AcTO3EOVcRwEwACci6Lkz03o+Q5yQAeA1AWwB/AujjvV8Vbr0H1ckf9Lqo5tKyBteUV2/PWoB6GFRzyUiXbJ/k24M+F4A1FwC4dWzo8f2zxtCyhcs4D6yay8jk8D4Zv4t9Lqq5vFGB8/3D5rHOkRFIa+fSXJpz3rjSX3hbd4/jQ6yaS/PerGvcNZt1jSqteGyquWy5l3t2BX0yd54qfaH6suaSOvoDilWzSZcPmGouDU/nnHjvudwz7vmJvK9ytb2/PNBrKiN8DWtB53O5/jPu96WaS/V/f0vxplXDKB7zJfdlC/LXVzpSrD6ZVy/jc1s1l5jjT6D4giNYswnncwGA9DWsbQV9Mr/M4d54qrl0rMwaovpk4ruz/qmay0vf8ZQPgyp/SnGdDP6salosbV1oextFd1Ofi2ouMRX5O6Za1YJpY6WcvQA6e+93OucqAPjSOfeR9z54ov4VwBbv/THOub4A/gWgz4FWtp9DqzmPYRhGIQleXA4FDqYPxmexv8KiQvafZul6AHg1+/FEAF2y5wPLE7vAGIZhlEJ8FP+CdpHsv8G6vex2YPMBbAQwLVAlvJ8jAKwBsnR4ANsQLHE9APlKkTnnVgHYASADQLr3vp1zrjeAfwJIRlabmO/zsy7DMAzj4BK0i4R5TgaA1s656gDedc619N4XaKZipSAazJne+2DCfxGAngCez+8Kgv3F1Oeimov6ZOK/5np29bmo5qI+mbo9OH+vXpeg7qIemRjpLaY+F9Vc1CezciFXyKvPRTUX9clcsiekk6jPRTUX9cn0T+OcuPpccmku4pOJO4/3U9DnAuTuTRb0yST0PY+W7X3zI4oj+WTWp/A0xupzUc1FfTK1OD2fu+19UHeR86fCNexLKeh8LtpbTH0uqrmoT+aJw/PWYFRzUZ+M9hZTn4tqLuqTyZjxBcVBnwuQuzdZ0CcTyeeimov6ZOqkh/e5qOaiPplrpL+Yel2Cuot6ZJx8G6rPRTUX9clEm5IySHrvtzrnpgM4F1nf8/tZC6AhgN+cc3HIahH25wFWkUOhU2Te+xTv/bLCvt4wDONgoqJ+aedgajDOucOy71zgnEtElrdxqTxtMoCrsh9fAuBzH6Fbcn4vMB7AJ865uQfK3YUjmPv7JHVFQV5qGIZhHBzqAZjunFsIYA6yNJgpzrn7nHPds5/zIoBazrkVAG4DcFce68ohvymy07z3a51zdQBMc84t9d7PzM8Lg7m/d+peXha8Q4ZhGMXOweyC7L1fCODEA/z/8MDjPQB6F2S9+brAeO/XZv+70Tn3LoAOyDLlFIjgnC7aW2zja6sobjCQ8/O/vcSzAWSsZN1i2jusNbRvyPn4XHO6SH+xEQHdpbdoLBNEg2nVjOeowG7O81a6oz/FR97LvhotvFi2gF9/VzK/t4dXhnSOYWIP0rljkq/kdb+zin0sd0kvsVw+F9Fc1OeiGs0u8L5Zuzqkk+we9hktq92Wb5jV56Kay9R01lyuEY0mtjofp/ienK/fMn0hxUNEJ/GBlHokzUU1Gv3wT3iF92Of19irEuwtBuT2uajmcuvv03Mea1+z9A8nUaz6zu7/jqW4z6nSS6w6n0SquWydtYNi1WjGDZlHcVDz2f4Nnw/aWyxzJ//GPPKhThT/OpTHcvhb7OmZMIv3UyfReDbM5JOqe8eQF+unmXw+qeby0lzRBNvw+fbBPF5+UVXWe6KNP/RbkeWrm3Il51yV/Y8BnA0WfgzDMAwjF/m5gzkcWSVr+5//hvd+qnPuYgBPAzgMwAfOufne+3PCrMcwDMPIJyU5UVi0iHiB8d7/AuCEA/y/9SIzDMMoJsrFBSaa1Bt4ZM5j9anUu4rzmxvHcv6z3pk8VPXJqK8m6LkBgHN6cDZQvSq9dod0jrdFc7n/SM61qk/mKfHUZEqfs0onsy6iXpVjkvm9rEjh9Q87IaSbqEfm/ivZC6KazE2NWHNR38rwEadQrD4X1Vx2zOacd/80Pi5Bn839bbln1aa5/NxuseHnc1HNRX0zNzzdimKdT2Zwf97vOsfL9Z/ekvM4kuaivcq0r5l6bCJ5VSIR1F3UI7Nr8QSK1XOj87eobyUuzHwuQG7NReeX6XcDH4egz2ZQV9Yn1ady/ZMtKd7zBh+zw09k4UF9MuqrUd9N79bbKN44P3TMj+7A/Q7V56Kay8sL+H0OjuCxuRKGclAvMIZhGEb+KAslt3aBMQzDKIWU5FTH0SJfRkvn3Crn3I/OufnOue9l2d+dc945Vzuv1xuGYRjlDxfB6Z/1pKxml+2kFxmccw0BvACgOYC2ulxZ1bprzsa0l9j6V3nukKBeAwBTH9lFsWouu3fy/O31zuTL/+0fcq5W+4m1aRPSKrS3GHazf0L7mN08gDUX1Wx+Xs7XXu0npgXvqtkEdRXtY3bPOH7fw5qx5pLQhLWAuIs5v37ftV9TfJP4ZB5ex3Pf9E/j+WSOaJR3L7Nn3q1Gy1RzSUwK30vMi3agvc2euYl9LtrLLKEZb7/CEJ7T5dmzRuU81l5io0SvyTWXTB/Wl1SzeeXNSghHpDldgl4X9blUasFeN/XJxFRln4vO57LvZ+nLJ5rN+K/Ya6K9zbZM4v0c1Gz2vDqZlmkvsWdvYYfDoC6s2bw1jTUb1VyqVObjpL3N1OsS1F02LeV9rL3EJs/k932RaC6jP2LNRXuZ1Xr/i6jeczx+ZL+oZcn+9uuYErkfKmq7/scB3IGykS40DMMoNRzMXmTFRaF7kTnnegBY671fEO6FwV5kb/z5WxGHaxiGYRwqFLoXGYC7keXqD0uwF1kwRWYYhmHkTVn4ssyXBkMvcO6fyJp47CYAqdn/3QDAOmRNPLYhj5ci9dGBORtTn0udKxtTrD6Z+ETOz6vPpWvPLfx68cms2sy5WfW6/GdQKOeuHhntLaY+lx2zwvtkHmjCfdDU53LCAB6r+mSCmoz6XJq05Np+9ck8+PxpFKvPZcdCzkOrT6ZnGvs3dD6ZkRelUhz0yWzewDqE+lyuO4P1HvXJ1B/ZhWKdTyZjO+8L9cnc9OkQitXr4hJDx1w9MjUi+Fyu7cOaoPpkXpnA88Uog95gHU99MjFHhvwi6nNJX8XHXH0yW29uR3Gu+Vy2so6mPpmEDo0p1vlkrhzVhuKgT6bWXfybU30uMZX5GKtPplNmBJ/L6azVqk/msGQ+H4Nel6YdWXsKemQAoG5H/hyqz+WCCD6Z26OsczzSKHoazB2rS0aDiXgHk91/LMZ7vyPQi+w+732dwHNW4QBFAIZhGEbhKC9O/gP2IivWURmGYRiHPIXuRSbPaRytARmGYRjlVIMpCu8FJhwbn8A55+fOYF9BhZO5Z1FMW/YNZH49jeLrHmcd5KH6nKeuegrnxFPGcQFd/SNDPYxqDpIc82ie/yJlFedmExzfzLY8mTWX+GPZ17J0LD//6ThO8D/ShDONu7eEdI8l63ldJ3VgHaPiKY0pnvEE59tPai3z6uzl1OzypeH9sm3+ys93dXlfbJ+wJOfxxJUNaFn/zjzW+LO5D9qeSbMoXj2bj1mTjqx7rJ6VRPF36ex72SY1kvVYxkP3SwI5+RhJUWfy5yL+Rp68L33Mfyl+8yX+rXZ5GJ0CAD5YW59i9Zrsnb0qNLT48PO5qM+l+lPkhc7lk0kcdjfFaa/we4k9m/UhV4n36yt9WQsL0qs56xR//MIaynd7eN6mU6vyuX54O/ZGvfM5a4LdW/P6kwZfQPGrN7I36vKzQ365tI38WRgjGkonx99BjU/heXF8Op8Tfyzi/X5sytSo6hwjG10RtS/nYavHHpI+GMMwDMM4IPkqU84W8Xcgq3os3Xvfzjk3HkCz7KdUB7DVe9+6WEZpGIZRzigvIv9+zgxWiXnv++x/7Jz7N4BtB3xVgGBarM9ebttQ4WRuDZP2DbeUqLBX2rWc0pXih8bzbf7QddyK+8ZxMrXwZRzHJodSGptfCJ8SS27M6bik+lzGuegbbq+SNIdv+5tfwTeOj8zmNMEdKzlN1Wc3t4M5rl6ojFlTYnu+XkXxSXLJ/24+p2aqOh7bsc15LJVP5fc+70W+a2967BKKq/Y+Lufx/7d35uFRVFkbf09CCIQQAgqIigSQRUTZYRABldFxRVwQRJRNcUMdnXH5BkfUERdGx8EFlWERRJBFYXABZQTBjU1AAQkiirIoiCwBAoEk5/ujKuk6pyu9JJ2Q5fx4+qFv1+3bt6uqU93nPe+5gwEcW7kxv/2GCnfckClDYlV6dRXthlVkGZsfl8i05xU5MoTWqZI8BdcelaGdX/TZ7g2LhQmJHX3padGe8Y5Mue47WMbfdEhMl8Dvv1b6k3UqsF7m2JtKHK7cvg6JBaUxnyI/KwkDh4k275chXj4U9qOdz9vp9dGtcsAyULuRTDvusk2+z88zVEh2oWxercOq18qQWObY90W730UyfDj1I2ln8EaE+6tSL/HJcp9v+UIu4Z2UJENstVvK/R5ryoMGU+RqyuSkl10H4IKiT8cIh/fiUtrxXlyMwqN9KqUZ78WltJFVDqoTlzUKXSrGQ1cAO5l5U2ynZhiGUXEpD7XICl0qhpmXuNuuBzCtoCe6F6ShANChVmucnpxWlPkahmFUCMrDejARXWCYebv7/y4img2gI4AlRFQJwNUA2oV4bn4tsv0DejDgpAJqzWXBo1LXmJ4o4+sJi1UastJcao2WceRXVBrz3rdkHNqruQBAzobAD7AdP8vYvU5D1ppL5g6VYsryh+F4nYasNJdqHWXq8V2b5PgNO8r+Xt3lt+ky7fjbX6TO0e4MWblHay4ZLPWd+EQZ+dVpyFpz2aSWImj7a+A46TRkrbm8uVSlMSdJzUWnIWvNpUO8TCtt0FWWCWlYU0WxVSrya+8EjrNOYe6VKs8vrblcd7UsO1Kpv9Rs3p/wqmhrzSXuLGktu+wUqdl4U5G15hKu3H7Vv8s0ZK25BKUxb39StN/6TJat1wwYL4+rN4152nUyhVlrLjVPlfaETj/IkFrd5lKz0ZrL0VmyTNPcNTLVuKcq56LDYt7TO6GO1GtyMuRnQ2sumZmyP1WSurARTNgQGRFVI6LqeffhlIrJU+D/CCCdma1McgmhRX3DMCIjsYyp5rngmN2OF0UtFdMXIcJjhmEYRuEoY9dDX4pUKoaZB8Z6QoZhGEb5oMhpytHgLf+ifS5ac9E+mQ6NZD689rlozUX7ZGoly5IV2uvi1V20R+boZukL0D4Xrblon8yoJBmH1j4XrblonwylpuXf1z4XrbnoNGZd+kX7XLTmon0y7Q4Xl9QCewAAIABJREFU7HMBpOYCSJ9My66hfS5ac9E+GR3S0D4Xrblon0zjq9QAyuvi1V20R0aXftE+F625aJ9M/zulrqZ9Llpz0T6ZnE8WB6atfC5ac9FpzLr0i/a5aM1F+2Reqhtag9GlY7w+GV36RftctOaifTK69Iv2uWjNRftkXl8st2uvi1d30R6ZC5Ll3LXPRWsu2icTa5d52UlOL5gSvcAYhmEcL7SoX9o5ntpJrLBaZIZhGEaxENEFhohSiWgWEaUT0QYi6kxEtYhoARFtcv+vGX4kwzAMIxI4hrfjRaQhstEA5jPztURUGUASgL8B+JiZnyaihwA8BODBkKN46oklDrtDbEpYPFW0teby1+3Sh9AnSy53emyVLCSQkCi3fz5C+kHOSJNTO5wT2BVac/EuuwoAbfvL/Pj9n8uy3rrc/6K/ySzux5ROUr2+HM+ruQDAhpcDMe7mN8jY/vS1Mnd/VE05FtQC1rq2mF7u+fyH5feERU/I53d6b72c29dyPK/GE1RuX9UWm7FIai7dlMayJEfG+ht0l5pLbqbUNZo9KXNRspetFm1dX+zIOwGvyu0DpcbyyuvyozFnnGz3zHgGoTi2Ri753buDjKhXqiO9LV7NBQD2fRo4p7Q+M/k2qR/qUv9x7TqJtq4tpn0uWnMZtnORaOvaZtmzZ4i2V+PRmovWSLL3yP2sNZcf75P7oXYjhGTnSunj0iX3vfXFcjOzxTLJff4gP5fTlS+rX8pO0Z64QGo2Wt+JNeVBg4nEB1MDQDcA4wGAmY8y8z4AVwKY5HabBKBXcU3SMAyjqHgvLkbJEMkvmIYAfgMwkYhaAfgKwD0A6jJz3teTX+H4ZQzDMIwYUFFE/koA2gJ4hZnbADgEJxyWDzvLYvruDSIaSkQriWjlhC++9etiGIZhKCqKBrMNwDZmXua2Z8G5wOwkonrM/AsR1QOwy+/J3lpkR1a8nf9e9ZLHeolj7XPRmotecvlCtcSy9tmckSZ3s17jpd05AbFC+1y05qKXW275uNRctMemRT0ZI9fLHnfvI9+b9rp4dRe93PKoJtJXoD02Y9QSx3o9l3bDpa/Fu+QxAHRqHdon0/4MGWP3+m7OvlDOTftUtK9FL3l8Yw85tvbJDHm6oWjrJZeTRkoBSXtV+g4OeKO076WeCoAHrSWjPDW6zpn2pgQtexxmTRev7qLXlul/p/R6aI/NgOkF+1QiIdx6ModWvS7aXt/N1RdIX4te8rjvmLNFW/tctOail1wO57O5qtt20dZelQYdAxqNXvLYu7wyEOyT6a+Wa9ZLLt8DQxP2Fwwz/wpgKxHlrV7ZA8C3AOYCGOA+NgDAf4tlhoZhGDHAe3EpC1Skcv13AXjTzSD7AcAgOBenGUQ0BMBPcBYdMwzDMGIAlwMNJtJy/WsAtPfZ1CO20zEMwzDKC+To8yXDgLRr8l9May6pfZqJtq4llvXSGNFOUJrLzU9J34GuZVYT0i/SsrOUjL764qT8+7qWGKlAova5rHtks2jrWmZZG+X6Hbrk/uJ/yxpHup7YiH0BP8mohjIGvXWT9K3oOmYHv5T7WdcS+2qk3A9tw2g2upbZyg0yxt6pdWB9mtnrZYxaay65LF+rYbdDcvsRqS3oWmbjH/pRtG9UnotZH6v4v6onlrsvoBdoj8zRMaF9Lt61ZIDg9WSuuEKeQ1pzyd4qwzVas5m5IrDvtM9l7xypBWifzKTbpQaoCbWeCxDa5wIA1doOFG2vZpOTodZOUj6Xt+74RrS1T0ZrNlpzyT4q94WuZbZ3m/zce+uJaV9W2jnSvzbpc+kH0j6XmaoOmtZsUsYviOkSYcPS+sTsj/NLW6Yfl+XLrBaZYRhGKaSipCkbhmEYRtRE9AuGiFIBjAPQEk5a9WAAl8Jx8+fCSVEeyMw7ChzEMAzDiJiy//slQg2GiCYB+JSZx3lqkeUyc4a7/W4ALZj5tlDj7Djn/PwX0z6X12b2E23tk4GqLaZ9Lp98KHPWtU/mpbZSi9Bel0YNArqH9sic+6SMvWqfS40uMtde+2RaDJOxX+1zObBVlhHXPpl2LQJxau1zeVmFabVPpsWD0reifS7Vu8r3Gk5z0evLnNFKag1en0z9qjI+rn0u150v4+/aJ6Nri2mfS1ySjMdrn8ygQVIY0V6X/gsDp6v2yGhfi2buDPletE/mtutkfF/7XLJ3hV7jJaH1afn3tc/lplelBqh9Mu/tkMdcM/CtS0Rb+2TiTm4i2np9meztUj/y+mQOzJRuEO1zqVRL7qhwmov2uVzeQuoi2idTr63U8bxelzqtQ6/nUv9suR+0z2XQhaF9MrdtnRJTnePWtN4xu8a8tmVm6dRgPLXIBgJOLTIAqqIiqqF8XHANwzCMGBGJBuOtRbaaiMYRUTUAIKKRRLQVwA0AHvF7srdUzJSdFkEzDMOIhPJgtCxSLTJmHs7M9QG8CWCY35OZeSwzt2fm9v3rhv7pbhiGYThwDP8dL8JqMER0EoClzJzmtrsCeIiZL/P0OQ3AB8zc0n8Uh2kn35D/Yj1Vbv+nL8ioW6smMt65bpPUTDrfIDWWY8pXEF9LajYZa6QPptZN0g+y6z8b8+9XSZF9V3wn48Q9HpBx3yXPyHh7mzPkIizDt8g48iMnyhpd8QnyGFSqImPu8PhFUv4mCya8P+Bz0e6cJnWN+3dIreCmI3K/dOgo+yfdd6N87R/SZTtZxq1ve1D6GkbWDby3aqeqelxKM8naITWS6te3E+2MN7+SL32ujHlPmSi1q/5D5X58dZwMOw/uINf/mLEssP5Hv15Sozu2Xcby926SPpaMDNlu0EZ6nX7bmCTaddtIzeX7T+VxWQ15TmV4vvrd3EOeT7kH5X6r0u9i0T44Zp6ci9Iplh2R3qlwukc0a7pU7z1abNN1zRKanSTace07iDaOyDV/4tvL9zau678RCvXJQbLnFLysudRvkvtILSs3XfrZ4lufKdrjh8vnD7pFnm/VHnkzpjrHzWnXxuzKMG7LrOOiwRS6FhkReZXAKwGkBz3ZMAzDKBTlIURWlFpk49yLTi6cWmQhM8gMwzCMyKnotciuif10DMMwjPJCiZaK8eouc1Vu/3lKc/l6k4y3n3uDjM1++aasOdSpl8xxz9kj27VukvLQnsnSD1IlJfBD8kiGXOdbay4fj5L+jo5Kc1m9QcaZH6kvvSKP75Yx8JEN5fbsI1KrqPXYtfn3M56UdaI6p8ko55dbpF50U67cz5OryP3SXWkumf96Q7ST+l8g2jgo9Sav5gIAw3cG3tsLTaUukZsp56I1lwPTQmsuBz+T50j/oXIN9SljlebSSWouE1bI/kM9usvUOdKX9cdUOdeaTaSGUjNXtn9anSraDbtITXDnaqnZaM2lDeQ51aRHoD3uY3k+3TFanstHps4Xba256HpdXbbJ9xZOc4lmTZdwa8kceOcK0c5duUK0Ewb+RbRzVsr3Fo5kFQ866Pl4aM3l4HTpZ6s+QNZoy1mzXrQH3SL1x4n/kefbMN882sJzPENbscJqkRmGYZRCckuwEHFxEVEtMiJKJaJZRJRORBuIqLP7+F3uY+uJaFTxTtUwDMMoS0T6C2Y0gPnMfG1eqRgiOh9O9lgrZs4iojqhhzAMwzAipez/fonMB1MDwBoAjdjTmYhmABjLzP+L9MXm1+2b/3ztc/nkB2nC1D6Zo2vkei/a57JsjoyBa5/M7oWyrb0u2VmBH3PaI7PkcekT0D6X5emhfTJ7Z8r8eu1zGf6jrAemfTLLtwdi8NrnUilRRmq1TybzxemirX0uQwbKWlHaJzOtqtxPWnM5YaiMa3t9MgfGLhSbtOZyZJ6MgWufzFuL5H7VPpeDi6TGEs4no70uY+cGdBftkdmyQvpUtM9F//bXPpm9+6RGqDWXbslyzR/tk+HswHvVPpcx98g6fNon885HUrPRPpeap8rPgtZs6jaXmk00a7pUaiy/Z2qfS/WrnxNtrdlM/UyuyaIZ9Hxz0dY+mX1DpIbj1V3G/0NqnVqvOaSOqfa58D6pP2qfTNWbnoqp16Rfg6tido2Z+tPs0umDQcGlYpoC6EpEy4hoMRF1CD2MYRiGUZEoSqmYSgBqAfgDgPsBzCCioKuktxbZB4c3682GYRiGD+WhVEwkF5htALYx8zK3PQvOBWcbgHfYYTmcrLoT9ZO9tcgurdo4VvM2DMMo11QIJz8z/0pEW4moGTNvhFsqBsBmAOcDWERETQFUBrA7xFBCd9E+l553y/i79slcdoXUBrTPRWsu2idzVjPpS9Belzq3NMu/rz0ybc6Qh0j7XHo8GM4nI9+L9rlozUX7ZB7z6C7a53LZpC6irX0yKSNC+1xuOiLH0z6ZUXXlGhlenwsAvKpqk3l9MuF8Lokny9NP+2S05qJ9Lv0HhfbJ9Osl15fRXhevT0Z7ZAa1kXWntM8lJUX5YpRPZuEKqatpn4vWXLRP5rSnuuff1z6Xm3vI/aZ9MleEqS3W6Qd5vmmfjNZc9JouV19Q8JoufXV9L+VzCeeTeaFuaA1Gay7aJxPK65KcK9d3Oai+Xg9Wmov2uQwZKTUX7ZPBTf5zrsgUpVTMIQATiGgdnPVhBnC4jAHDMAwjInLLQR5ZUUrFAED/2E7HMAzDACpQLbJY4S25r0u/5Pwsf7Zf3EWGat5/V/501mnMGfNkSCMHslx6uJL7B2YGUj/XbZEhh5Zpu0S72yMy5JAxXaaNnnWSDL95y+0DsvQLAOwZMUu0L86S4cARv3hCO4lZIqSW8ZRMQz52WO4X7Noumkm9ZLLftLU/irYOia3bItNOj6kQWuYUmYrsTYPOXfiR2MYqGBwuDTl7k0y/va6TDKEd+kL2T7lBhuR2/FuGMHp3kOfIlhWBsJcOiVElecwWQoYCk/fJdr9TZAp0howGitIvAFC5559Eu062DPX8/H+LPa2qIqQ2Y4E8P7vr8Ft7mVoOeYjCpiH/eN9i0a7dSD4/5DLHqty+Lv0ySZXb1yGxu3cuEm0dUstZt0S0c9fJYxzXKE20k1oF0s17t8rA6zMDy5frkNiM1+QxZxVCOzDpC9HW4TgjGCsVU8bQeo1R/tF6jVE4vBeXsoDVIjMMwzCKhfIgaRe6FhkRtSKiL4loLRG9S0Rl6+uBYRiGAQAgoglEtMtN2vLbfh4R7SeiNe4totrRha5FBmABgL8y82IiGgzHbPn3UIN4U4l1GnH7bqGXPNaai05jrpkjU1Z1KZoaV58u2jqVOIEDcW1dCqZ679Cl/rOzpOaiy9Do8i06lfjLLTIOrcvBdDwWiA3rFOZH42XsPyFJLRqryuvrJY9DldsHgjUXXUpGl57xpkFny10cVMql//mh05C15qJLyVS/pJVo6zTomk1EM6ici7f8i05D1pqLLiVTqZ7U+HQK9M0XynNIpxIPPig1F10Opu6MD/Pv6xRmnfKsy9CsXSjbuvx+5WtDpyFrzUWXkgm1xPINYdKIwxEujXl/M5n2rEvR5Cz8RLS95VyGtAZytwRKTuk0ZK256FIy4cr9Jw1FTCnhLLLXAbwEYHKIPp8y8+XRDBr2F4xbi6wbgPEAwMxHmXkfnFIxeYrbAtgCZIZhlGK8F5eyQEkaLZl5CYA9YTtGSVFqka2HU00ZAHoDqF/QAIZhGEZ0lMJSMZ2J6GsimkdEZ4bvXrRaZIMB3EFEXwGoDsdsGYS3FtmEb36K6F0YhmEYscP7d9i9RRvQWwWgATO3AvAigDkRvW4E5fpPArCUmdPcdlcADzHzZZ4+TQFMYeaOocbKuPVP+S+mS72sXCLj87r0y4HlUkvQpWb2xsv4vNZsDi6QP491uZdjnjqd2iOTMWeTaOsyM5USpVagy/3vnSI1G13upWsz6VUJ0mw8XhddZkZ4ZBCcxvz33VJbCCq3f1930daaTeYcGfPWmsvivnK1Bm+pme45slSL9rngmNSqdKkXUl9/dOmZo4u/Fm1daiZcyf29mwPahi71MnOF/EGuS/1n/yL9HrrUTLhljScsk/11yX2v10VrLtWT5WdHpzHnZMigiPa5HJ31oWhrX4tGay7ZR+U56C01MzM9dCAjXLn9IJ+LKjVT40m5XWs2U1S5f11yPy7ttMBrqVIvmV9LD5jWXKY/Jsv961Izd22dEtOS+JeedmnMfnp88PMHYedGRGkA3mPmlmG6goi2AGjPzCHLg4X9BcPMvwLYSkR5xbp6APg2b4ExIooD8DCAV8ONZRiGcbzwXlzKAswcs1tRIaKT8qrlE1FHONeO30M/q2i1yG4iojvd7e8AmBj1rA3DMIzjDhFNA3AegBOJaBuAEQASAICZXwVwLYDbiSgbwGEAfSOpPVmUWmSj3ZthGIYRY0rSyc/M14fZ/hKcNOaoKFEnv1d30T6XcOX2dW0x7XNJuUTGfoN9MlJzCfK69AyYJrRHRtcWC/K59Antk1n6s4xxa59LOJ9MdlYg5q2XWw5X6n9kXakdBJXbD+OTCeVzAYDJVaQG5PXJdL0ztM+llzqG2icT10DqFNrnkpAqx9M+Ga25aK9Lg3aB7dojozUX7XM5GKdK/yufzJGpsq19LjcnS+1K+2S89cW0z6V3G6kVBJX6f+Z80dY+l7lr5GdF+2R2rpTnuy73f3kLWbdN+2RCEa7cflBtMeVz2X2N/NwG+WQelkuvS6/LVgzsHfDb6SWPk5tLLUv7XA7Gyf2mfTKxpjwUu4zIyW8YhlHW8V5cjJLBapEZhmGUQirEejBu9pi3JnwjAI8AOAXAFXD8L5sBDHId/oZhGEYRKQ/FLsP6YERnongA2wF0AtAMwEJmziaiZwCAmR8M9fz5dfvmv1imMjlkxsn2lT1lerWe5/5V2aL93c8yTtz5MRnT3jf5G9Fev1muc9KmdSC2W234ILHt4D9kglzuMRn7/8uPMh4/6hSpi3yzWWoL06tKDefmI3K8tqNkES3O9Hgu1KIqO16UMeiTb5apmLeOlpmE/2wgtQWKl/s1J0seh/hE+XopD98g2nv/Pk200zcHlgpOr1xZbBtwh9RI4tLSRDt3o3wvWat3iHbVfheI9uS/SH9Sjsr0T1Cn9gWp0sfwXkZgrufkHBLbGrWW+ylp0IVysFy5X7Jmy3VMcg7K7RO/lvH7OqpkXK/L5dx+XRJ4M3VaS9/LpiVSS2rcUc51xkrl4blIal3HdklP9BQ1t+4kQ0lp50idbtcaqZ/WbhnQLqYslbqZ1ikuay71G+01oSpST9JeFV1yX/tcajzxiWhrn0xCy8B7PbpKeuOq9JHnV/y5svpV5r3DEIrU6Yti6oPpcepFMbvCfLzto5jOLVKiDZH1ALCZmX8C4LXlL4WTxmYYhlEq8V5cygLlIUQWrcjfF8A0n8cHA5jn9wRviYIPDm+Odn6GYRgVklJYiyxqIr7AuCbLngBmqseHA8gG8Kbf85h5LDO3Z+b2l1ZtXJS5GoZhGGWIaEJklwBYxcz5AV0iGgjgcgA9InF1enWXJKUlXNhTxpH/O1dqKt1Pkbn6NdrKqXceIjWXL0fIfPlcSM3lzMa7RNuruxwaGVpziVPBfa25PLBdek0SlObS57D0GbT9Z0PRXvWA1BbajPD4aJR2pTWXHeNkXPmfDWT8/v6fpF70jJp7OM0l4wn5PSJd6UvNGwe0hE5XyLlNGiOFhwFDpOYS10zGy6ueIetWHZ4qF5fPIRnyiA+juSzcV1u0z+GA7vJFvKyb1nKQrHuWOXGBfK0UeRwSr5Lek5f/Kt/boFZSe0g4WWoNc96Tc+vZLVCfTmseWnPZvFwe034XS81l6kfyGGWpaHx/Nbf4ZPnetnwhvVENOkqN5rd1gfeiNRddryvsmiqtZP047VUZlCrPb72mS6j1ZH47Kk+Qym3l+ak1l5zP3sbxJLcciPzRXGCuhyc8RkQXA3gAQHdmzizwWYZhGEbUlP3LS+RLJlcDcCGcmmN5vASnTP8CdwlNK3ZpGIZh5BNpLbJDAE5Qj51eQHfDMAyjiJSHLLISdfJ7vS5ac/l8tsztTyIZzA3yuSjNJXt1umg3PU36ZKrWkm3tddn70KT8++s3y7EnVZEaitZcEqrJuT7XUL6379Pl3LXmsvj+H0V7elV5Yg33eF205oIkWaPt5LtkDPrADOn/0ZrLg0ov0j6ZYJ+LjOef2ULVhPNqNrvkOjdac5k0XvpiBtyxRbSnqHpyWnPp21ZqB9onM/Z+0RSaCwCsoYDucnmK1GuQK/ej1lz0mitQPi7tc9Gay7Edsu7VLrkrhNdFay6702WdvibdpL/52C7l0VGaS6L6u5VQR/qVfv1C1XRTmstPy6UXxeuTuWxXaJ/L+H/I/ZycK4/pIXmKBGkuuuT+wN7SJ0MnyPG8ukvtd6W2ubuy3DHHwvhcEq+ROpvWbGJNebjAWC0ywzAMo1iwWmSGYRilkPJQKqYotchSAdwCIO8379+Y+YOYz9AwDKMCUh5CZGEvMMy8EUBrQNQimw1nVcvnmfnZSF/MW19M+1y05qJ9Mrq2mPa5aM1F+2QSbpDrmmivi1d30R6ZUZBon4vWXLRPRtcW0z4Xrblon8zJdwfiztrnojUX7ZMJV1tMay7aJzPoiNyPXp8LENonk9xdrs2hfS5ac9E+GSjtQPtctOaifTLn5Mj3or0uXt1Fe2R6q9pi2ueiNRftk+l1eWifi9ZctE+GPLtd+1y05qJ9MvMOyc+W9rlozUX7ZPr8Qa5l4/W5AMG1ybw+mabDpZla+1y05qJ9MoNVbTHtc9Gai/bJHPnvl6Lt9bpozUWvJbP7KvVZUoT1yVwn52IUoRaZuzyzYRiGUQxUxAXHdC2yYUT0DRFNIKKafk/w1iKb+O1Wvy6GYRiGgpljdjteFKUW2SsAGsMJn/0C4Dm/53lrkQ1qUbaqmRqGYRiFJ+L1YIjoSgB3MvNFPtvSALzHzC31Ni9bO/TIfzGtkayYI30wWnOZ/6iM/WuNZne8HE+vJ5OZLn0HWie597uAj1T7XPTaMVqjqVxdzkV7bHi9XEte6yQ7XpAeHu11ufOFgE6i9ZnqJLUnrfdkTPxCtLVm8mU/uTa81ljC+WRys2Wo1KvxJEi7RJBmMvGvUovSGote30WvJ4Msua6J1ngyp3wi2npNl6y3P86/rzWWKffJ+Lyua7Znr/QfaX1nQBfpAdI6yf4M6WXRXpcJXwW+jGl95v1V8ouaXkvmh/nytbRmwtlyR2uNZeFBqeHo9WQmLpCajVfjSWwSupbY+OHyvej1XGa8Jg+61mj6NA2znkyqPF+9ukm49VxOnK00GVXXjMPUMqs2fHJMdYO29c6N2U+PVb98VurXg9G1yOoxc14FyqsArIvlxAzDMCoyFSJNGRC1yG71PDyKiFrDqcm2RW0zDMMwKjhFqUV2YwHdDcMwjCJSHnwwEWswsSDj1j/lv9j+VVI7OGGIlG+0z+WYMkVkKh3jwl6h15P505kydqvXeMk5Gmhrn0s4TebcmZeItvbYJF3VWrSh9CMclvqQ9rp8/3tgPtPV2jJjHpdxYu2xafumlMz0ei7aJ5O+Wfo1WjSTepP2yTzb6PcCx6t5z3lim/apzPxKagnhNBftkxk0Snoucjeq9WVayvi/9qokDeiRfz9L+V5+Xyu9Itonk6m0gXNyZJ2zGimHRbtOa7kuT+4ReQ5or0ujtoHzeeLXcj8NvURqLqHWkgGC13NJSpLaVe2W8vyLT5E6n/bJ9G4tP0tTPPMb+oQ6pmtUrbBUORftc9F/jfT6Mn1GyPeqfTapb4yRr+/xqmS9LY9xEOqYap/M/hE9RPvoKvk5rfn2JzHVOc4+qXPM/jh/8+uXx0WDsVpkhmEYRrFQlFIxnwB4FUAVOEsm38HMy4thjoZhGBWO8rCiZVQhMk+pmE4A/gOnVMw8IroUwAPMfF6o539ct0/+i+k05N/HyyQ0ncY8913501inIS+YI0MMOo15ShUZFtBhr/jKcj9405hVNC4oDfmz3vNEW6cxZ+6T4ZagkvtVZZqofsHVj8qQh05FvuORwE95ncZcGXI/6DTkcEsk63L9uv9ffxDSXFAa8zs/nhp4rjrVercLXW4/d4NM39ZpyBMf2CzaOqR27BsZwtCpyJmTPhZtbxrzuntkarlOQ05SoZtwacw6DTmuijzGOo15aYYM8XpTkcfOk58FncYcV0VGQ3QacmamPB91GvOkz08RbV1qZuYaGQbTacyTPSE1nYbM++Rr6TTmA5NkWr1OQ57+mNzPOo35xnPkZ0UTquT+obvuFNt0GnKNx+T5otOYU6ctimkY6sy6nWJ2hVm/c1mZCJHll4qBEy7NczrUALAjlhM7nmiPTGki1MWltOG9uJR2tEemNKF9LqWZyUqvKU2U9HouRvS1yLylYv4M4EMiehbOheqcWE7MMAyjIlMeQmRFKRVzO4B7mbk+gHsBjC/gefm1yN47vNmvi2EYhqHgGP47XhS6VAwR7QeQysxMTmnl/cycEmqMzFfvyX8xnYbcoZcsQa7TmFPbSW1BpyGfmCP76zTmPzT6RbR1KvLzTQPptjqFOeXxgaKt05CPHpCvpdOYuzwod0tQyf27m4u2TmOmasmi7U1FPsDyR6hOYx55soz96zTkzlP/KNo6jVmj05rjKsnzx5vGfG6O1CF0GvKgZ2W4T6cxV+17nmjrNGQkSi0hKI35ORkj16nIidcE0k51CvOUtVJn0GnItWpmirZOY76mg9QtdBpyuDTmuJTAe9NpyJeppaJ1GnPPalKf1GnIVEkeCJ3GXP/s/aI9RY0/6EKpuXjTmG8MU25/yMjQacxxzUOX+5/+nXy+TmO+6tzQGkzS8y8FXluV28/d9L1o6zRkClPuP/vo9pjqHM3rdIjZlSF914qyVSoGjubSHU422QUANvk8x4gx2udiGEZkeC8uZYHyECIrSqmYWwCMJqJKAI4AGBr76RmGYVRMysN6MEUpFfMZgHbFMSnDMAyj7BNtFpnImCm6AAANyUlEQVRhGIZRApSHEFlMV02LYnW1ocXVvzjHLstzsbnbXMry3EvTXArTvzC3hie05ljdinuuBd2OVy2yaPWaaPoX59jR9i9Nc4m2f2maS7T9bS6x6W9ziV3/ComFyAzDMEohrKuul0HsAmMYhlEKKQ/rwRyvENnYYuxfnGNH2780zSXa/qVpLtH2t7nEpr/NJXb9KyQluuCYYRiGERmn1TorZn+cf96zttQ7+Q3DMIwSwkJkhmEYhlEAJfILhoiaA7gSQN5KRtsBzGXmDT59OwJgZl5BRC0AXAwgnZk/KIm5GoZhlAbKg3xR7BcYInoQTqHMtwDkLal8KoBpRPQWMz/t6TsCwCUAKhHRAjgrZy4C8BARtWHmkWrsTgA2MHMGEVUF8BCAtgC+BfAkM8uysM5zGgG4GkB9ADkAvgMwlZkzYvm+SxtEVIeZd4XvaZQlyupxJaITmPn38D0rLuXByV8SIbIhADow89PMPMW9PQ2go7vNy7UAugDoBuBOAL2Y+R8A/gSgj8/YEwDk1U0fDWdlzWfcxybqzkR0N4BXAVQB0AFAIpwLzVIiOq8ob/J4QkTzVLuWup0AYDkR1SSiWj7PX0VEDxNRY72tgNerRES3EtF8IvrGvc0jotuIKMGnfxIRPUBE9xNRFSIaSERziWgUESWrvsOI6ET3/ulEtISI9hHRMiI6y2fsOCIaTETvE9HX7nt5y+94ElG8O+9/EFEXte1hn/5ne+4nuPtoLhE9SURJPv0bEdEEInqCiJKJ6D9EtI6IZhJRWsidGjzWPJ/HIj6uhTimNYjoaSJKJ6I9RPQ7EW1wH0v16X8SEb1CRC8T0QlE9CgRrSWiGURUT/V92nNM2xPRDwCWEdFPRNTdZ+xkInqciNYT0X4i+o2IlhLRwALmnkJETxHRG0TUT20b49P/YvW+x7vn8FQiqqv6tieiRUQ0hYjqE9ECd04riKhN6L1qlMQFJhfAyT6P13O3eclm5hxmzoSzNHMGADDzYZ++ABDHzHkLwbRn5j8z82fM/BiARj79bwFwCTM/AeCPAM5k5uFwwnDP686l6cQlorYF3NoBaK267wbwlee2Ek54cpV7X1MTQCqARUS0nIjuJSK/Y5bHG+5rPgrgUvf2GIBWAKb49H8dQF0ADQG8D6A9gH8CIACvqL63M3PegiajATzPzKkAHoTz5UAzHsBpAJ6C82v3Pfexh4noLtX3NThLTPwO4AUi+pdn29UFzDuPpwGcDuA5AFULmMvrAFYAOAhgKYB0OL/I58P5MiSI8pgC0R3XaI/pDAB7AZzHzLWY+QQA57uPzSjgvX4LYCuc/X4YznnwKYL3zWWeY/pPAH2Y+XQ4Fdqf8xn7TQA/wPli+RiAFwDcCOB8InrSp/9EOOfS2wD6EtHbRJTobvuDT3/vGM8B+AXAFXCO3Wuq7xgAo+Cct18AeI2Za8CJlgT9DYglXA4WHCv2WjRw/nh/D2AenNzxsXA+cN8DuFj1XQYgyb0f53m8BoBVPmPPBDDIvT8RzkUGAJoCWOHTfy2ARPd+TQArPdvW+fR/G84fll4A5rrtvOf7zWeV5/44AE8AaABnxc85qu9yOH98rofzIb3WfbwHgC99xs4BsBDOh1nfDqu+f3H38Vmex34McYy88+4K54Pzqzt2UM0lAN+FGCtoG4A17v/kjkue9jeq70bP/RVq2zc+Y+vnL3X/T4QTPvXtCyc8PBbAO27f1T5jr/a+BwAJBc3bp//PBW0rzDGN9rgW4phu9BunoG1h3usa1d4AoJL3+Hi2rfUZ+2vVXuH+HwdHj/U9vzzt4QA+h1MBPtznVD9Xt6M6prG81UlpxrG6Fec8Q92KXYNh5vlE1BROSMwr8q9g5hzVvRszZ7nP8/5iSQAwwGf4m+GsSfMwnG93XxLRVjh/sG/26T8OwAoiWgbnQ/cMABBRbQB7fPo3ZuZr3PtziGg4gIVE1DPkm3Zoz8x530KfJyI9/wRmnue+/jPMPAsAmPljInrWZ7wNAG5l5qAVx9z3nA8zP0dE093X3QpgBBDZ1xhm/hTAp+63/wvhhCa1qWwPEfUG8HbecSKiOAC94XzjLWhsJqIP2P10um09r1lE9DqAxwHMJqI/A5gNZ1G7nxHMMSJqzMybiagtgKPu2Fk+Y+cvE8nOL9+h5Oh+CwEkI5gaRHQVnD9sicx8LMS8ASDXPddrAEgiovbMvJKITgcQ79M/4mPqvm6hjmuEx/QnInoAwCRm3unOoS6AgXA+Txpv9GNyiG2Ac3H7gIieBjCfiEbDubBfAOfCrTlEROcy82fuZ22P+z5yicjPz5FIRHF55yIzjySi7QCWwP+41iGi++B8UUghIso7J33mfoSILoJzTJmIejHzHDe0p/9+xZTykKZcIllk7oFfGkG/rAIe3w3nAqIf3w9gIBGlwAm/VAKwLe8D4tN/NBH9D8AZAJ5j5nT38d/g6D6a0nTiPuozRh46FARm3gagt/sBXQAgSDPw8J1+wL34z3dvmr5wLs4vE1HeWtepcL4d9/Xpv5KIkpn5IDMPznuQHH3ggHrd4eTE2qcBaAzn18VQAHMA3OAz9v1wwkBZcI5/X3fs2nDCZXoeFzNz/nti5sfcY6pDdYBznPO+TCwlorrMvJOIToLP+QjgAQDvwgnn9gLwf+ToODXgXxzxUURxTN35Rnpcoz2mfeCEfRa7FxYGsBPOL/frfPr/13NM8/Ur92IqXpuZXySitQBuhxNdqASgCZxj+oTP2LcBGEdETQCsBzDYHbs2gJd9+r8L52L1P89rvk5EvwJ40af/fwDkrRM9CcCJAH5zj6u+4N0GJ0SWCydkd7v7BWg7nJC7EQJz8oeAiEYB+IiZ/6cevxjAi8zcRD0+Qg0xhpnzTtxRzHyTp28rBE7ce+F8+AbAPXGZ+Quf+TSH8ytwGTMf9M7H+0dT94VzwWrMzOv8+kY7tvt4Jzh/hDYDaA6gM4BvuYB0cvJPP98I4ANWJ6Hqe6bbd0OIsTvD0e+iTm0nosne41IM/d8D0JMjqFxIROfC+aW/jpk/iqB/Vzia0vJw/cON7R7PdGbeT04CQ15G5nr4ZGRSFBmcqm+kY3vn8mBBY7v97wYwm5n9fmn57YuI+0c7diw5MaVpzP4478747rg4+e0CU0iIaBAzB2WqxaK/X1/3RL8TTlilNYB7mPm/7rZVzNy2MH3dx+4CMCyK/vnp5HC+RXcE8Amc8MuHHJxOrvvnpZ8H9Y/B2AX2J6K5kBAcIXshADCzCH369Aecb8qx6r+cmTu692+Bc8xmA7gIwLvsSeEvoP8dcH4FBPUvxNjrAbRi5mwiGgvgEBzNsYf7+NVh+mcCmOXXvzjHdvvvd8fcDOeX70w3KuFLNP2jHTuW1KreJGZ/nPcc2HRcLjDHRfgpDzcowS+W/f36wklQSHbvp8HJGrrHba8ubN8i9I+HE57JAJDiPl4V/uJ3xP2LeezVcLLczoPzzf88OBlE3QF09xm72Pt77q8AUNu9Xw3+4nfE/Qsx9gbP/VVq25qi9C/OsT37PQ7OxXM8gN/ghAEHAKhelP7Rjh3LW83k0zlWt+KcZ6ib1SILARF9U9AmOGm3he4f7dhwsuoOAgAzbyHH5zGLiBq4zyls38L0z2Ynnp9JRCKdnIj8QkHR9C/OsdsBuAdOltH9zLyGiA4z82KfcUuifxwR1YTzB4zY/WbMzIeIKLuI/aMde53nl/PXFEhQaArgWBH7F+fY7tviXAAfAfiIHC9WXobmswBqF6F/tGPHDPcCV6axC0xo6sIR9nRmFMHJiS9K/2jH3klErZl5DQAw80EiuhyOv0IbEKPpW5j+R4koiR2/Urv8iRPVgL9fKZr+xTa2+4fieSKa6f6/EyE+A8XdH474/xWcY85EVI+ZfyHHfOp3YY+mf7RjR5uRGU3/4hwb+v2wk+03F8Bc8jHERtk/2rFjRnnIIjMNJgRENB7ARGb+zGfbVGbW5suI+xdi7FPhfFv/1ad/F2b+vDB9C9k/kX0y/shxa9dj5rWF7V+cY/v0uQxAF2b+W0F9SrK/53lJAOoy84+x7h+uL0WYkVmY/sU1NhE1ZeagrLkQ40bcP9qxY0mN5MYx++O8/+BmE/kNwzAMh5RqjWL2xznj0A+2HoxhGIbhYMUuDcMwDKMA7BeMYRhGKYTLgchvFxjDMIxSiIXIDMMwDKMA7BeMYRhGKaQ8ZPjaBcYwDKMUUh40GAuRGYZhGMWCXWAMwzBKIbEsOhkJRHQxEW0kou+J6CGf7YlENN3dvoyI0sKNaRcYwzCMUkhJXmCIKB7OYm6XAGgB4Hpy1lbyMgTAXmY+HcDzcFcEDoVdYAzDMIyOAL5n5h+Y+SiAtwBcqfpcCWcFUMBdn4fIdwnrfOwCYxiGUQrhGN4i4BQ41arz2OY+5tuHmbMB7AdwQqhBLYvMMAyjFJJ9dHvMClQS0VAAQz0PjWXmsbEavyDsAmMYhlHOcS8moS4o2wHU97RPdR/z67ONiCrBWXPo91CvayEywzAMYwWAJkTUkIgqA+gLZ2E1L3PhLBUNANcCWMhhMgjsF4xhGEYFh5mziWgYgA8BxAOYwMzriehxACuZeS6A8QDeIKLvAeyBcxEKiS04ZhiGYRQLFiIzDMMwigW7wBiGYRjFgl1gDMMwjGLBLjCGYRhGsWAXGMMwDKNYsAuMYRiGUSzYBcYwDMMoFuwCYxiGYRQL/w85JSgsfWHF2gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10c4cf9d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.subplots(figsize=(7, 7))\n",
    "sns.heatmap(dists, square=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Make crystal graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Spacegroup: 33\n",
      "Gamma: 90.0017\n",
      "CG Factor: 1.4\n",
      "Node count: 80\n",
      "Edge count: 176\n",
      "Ga_0    4   ['O_40', 'O_38', 'O_57', 'O_62']\n",
      "Al_1    4   ['O_39', 'O_41', 'O_56', 'O_63']\n",
      "Al_2    4   ['O_42', 'O_60', 'O_59', 'O_36']\n",
      "Ga_3    4   ['O_43', 'O_37', 'O_58', 'O_61']\n",
      "Ga_4    4   ['O_45', 'O_60', 'O_58', 'O_34']\n",
      "Al_5    4   ['O_59', 'O_44', 'O_35', 'O_61']\n",
      "Al_6    4   ['O_32', 'O_56', 'O_47', 'O_62']\n",
      "Al_7    4   ['O_33', 'O_57', 'O_46', 'O_63']\n",
      "Al_8    6   ['O_38', 'O_45', 'O_33', 'O_54', 'O_48', 'O_78']\n",
      "Ga_9    6   ['O_39', 'O_44', 'O_55', 'O_32', 'O_49', 'O_79']\n",
      "Ga_10   6   ['O_36', 'O_76', 'O_47', 'O_50', 'O_35', 'O_52']\n",
      "Al_11   6   ['O_37', 'O_77', 'O_46', 'O_51', 'O_53', 'O_34']\n",
      "Al_12   6   ['O_40', 'O_74', 'O_34', 'O_51', 'O_36', 'O_52']\n",
      "Ga_13   6   ['O_75', 'O_41', 'O_53', 'O_37', 'O_50', 'O_35']\n",
      "Al_14   6   ['O_42', 'O_38', 'O_72', 'O_54', 'O_32', 'O_49']\n",
      "Ga_15   6   ['O_55', 'O_43', 'O_39', 'O_73', 'O_33', 'O_48']\n",
      "Ga_16   6   ['O_64', 'O_42', 'O_38', 'O_73', 'O_58', 'O_48']\n",
      "Al_17   6   ['O_65', 'O_43', 'O_39', 'O_59', 'O_72', 'O_49']\n",
      "Ga_18   6   ['O_40', 'O_50', 'O_75', 'O_56', 'O_36', 'O_66']\n",
      "Al_19   6   ['O_37', 'O_67', 'O_74', 'O_41', 'O_57', 'O_51']\n",
      "Ga_20   6   ['O_76', 'O_46', 'O_62', 'O_34', 'O_68', 'O_52']\n",
      "Al_21   6   ['O_35', 'O_77', 'O_47', 'O_63', 'O_69', 'O_53']\n",
      "Al_22   6   ['O_54', 'O_60', 'O_44', 'O_70', 'O_32', 'O_78']\n",
      "Al_23   6   ['O_55', 'O_61', 'O_71', 'O_45', 'O_33', 'O_79']\n",
      "Ga_24   6   ['O_64', 'O_40', 'O_72', 'O_70', 'O_50', 'O_52']\n",
      "Al_25   6   ['O_65', 'O_41', 'O_73', 'O_71', 'O_51', 'O_53']\n",
      "Al_26   6   ['O_42', 'O_66', 'O_74', 'O_54', 'O_48', 'O_68']\n",
      "Al_27   6   ['O_43', 'O_75', 'O_67', 'O_55', 'O_49', 'O_69']\n",
      "Al_28   6   ['O_76', 'O_66', 'O_44', 'O_54', 'O_49', 'O_69']\n",
      "Ga_29   6   ['O_77', 'O_67', 'O_45', 'O_55', 'O_48', 'O_68']\n",
      "Al_30   6   ['O_64', 'O_46', 'O_71', 'O_52', 'O_51', 'O_78']\n",
      "Al_31   6   ['O_65', 'O_47', 'O_70', 'O_53', 'O_50', 'O_79']\n",
      "O_32    4   ['Ga_9', 'Al_22', 'Al_6', 'Al_14']\n",
      "O_33    4   ['Ga_15', 'Al_8', 'Al_7', 'Al_23']\n",
      "O_34    4   ['Ga_4', 'Al_12', 'Ga_20', 'Al_11']\n",
      "O_35    4   ['Al_21', 'Ga_10', 'Ga_13', 'Al_5']\n",
      "O_36    4   ['Al_2', 'Al_12', 'Ga_10', 'Ga_18']\n",
      "O_37    4   ['Al_11', 'Ga_13', 'Ga_3', 'Al_19']\n",
      "O_38    4   ['Al_8', 'Ga_16', 'Al_14', 'Ga_0']\n",
      "O_39    4   ['Ga_15', 'Al_1', 'Ga_9', 'Al_17']\n",
      "O_40    4   ['Ga_24', 'Al_12', 'Ga_0', 'Ga_18']\n",
      "O_41    4   ['Al_19', 'Al_1', 'Al_25', 'Ga_13']\n",
      "O_42    4   ['Al_2', 'Ga_16', 'Al_14', 'Al_26']\n",
      "O_43    4   ['Ga_15', 'Al_27', 'Ga_3', 'Al_17']\n",
      "O_44    4   ['Al_28', 'Al_22', 'Ga_9', 'Al_5']\n",
      "O_45    4   ['Ga_4', 'Al_8', 'Al_23', 'Ga_29']\n",
      "O_46    4   ['Al_11', 'Al_30', 'Ga_20', 'Al_7']\n",
      "O_47    4   ['Al_21', 'Al_31', 'Al_6', 'Ga_10']\n",
      "O_48    5   ['Ga_15', 'Al_8', 'Ga_16', 'Ga_29', 'Al_26']\n",
      "O_49    5   ['Al_28', 'Al_14', 'Ga_9', 'Al_27', 'Al_17']\n",
      "O_50    5   ['Ga_24', 'Al_31', 'Ga_10', 'Ga_13', 'Ga_18']\n",
      "O_51    5   ['Al_11', 'Al_12', 'Al_25', 'Al_30', 'Al_19']\n",
      "O_52    5   ['Ga_24', 'Al_12', 'Ga_20', 'Ga_10', 'Al_30']\n",
      "O_53    5   ['Al_21', 'Al_11', 'Al_31', 'Al_25', 'Ga_13']\n",
      "O_54    5   ['Al_28', 'Al_8', 'Al_22', 'Al_14', 'Al_26']\n",
      "O_55    5   ['Ga_15', 'Al_23', 'Ga_9', 'Ga_29', 'Al_27']\n",
      "O_56    3   ['Al_1', 'Al_6', 'Ga_18']\n",
      "O_57    3   ['Al_19', 'Ga_0', 'Al_7']\n",
      "O_58    3   ['Ga_4', 'Ga_16', 'Ga_3']\n",
      "O_59    3   ['Al_2', 'Al_17', 'Al_5']\n",
      "O_60    3   ['Al_2', 'Ga_4', 'Al_22']\n",
      "O_61    3   ['Al_23', 'Ga_3', 'Al_5']\n",
      "O_62    3   ['Ga_20', 'Al_6', 'Ga_0']\n",
      "O_63    3   ['Al_21', 'Al_1', 'Al_7']\n",
      "O_64    3   ['Ga_24', 'Al_30', 'Ga_16']\n",
      "O_65    3   ['Al_31', 'Al_25', 'Al_17']\n",
      "O_66    3   ['Al_28', 'Ga_18', 'Al_26']\n",
      "O_67    3   ['Al_19', 'Ga_29', 'Al_27']\n",
      "O_68    3   ['Ga_20', 'Ga_29', 'Al_26']\n",
      "O_69    3   ['Al_21', 'Al_28', 'Al_27']\n",
      "O_70    3   ['Ga_24', 'Al_22', 'Al_31']\n",
      "O_71    3   ['Al_23', 'Al_25', 'Al_30']\n",
      "O_72    3   ['Ga_24', 'Al_14', 'Al_17']\n",
      "O_73    3   ['Ga_15', 'Ga_16', 'Al_25']\n",
      "O_74    3   ['Al_19', 'Al_12', 'Al_26']\n",
      "O_75    3   ['Ga_18', 'Ga_13', 'Al_27']\n",
      "O_76    3   ['Al_28', 'Ga_20', 'Ga_10']\n",
      "O_77    3   ['Al_21', 'Al_11', 'Ga_29']\n",
      "O_78    3   ['Al_8', 'Al_22', 'Al_30']\n",
      "O_79    3   ['Al_23', 'Al_31', 'Ga_9']\n"
     ]
    }
   ],
   "source": [
    "cg_factor = get_factor(train.iloc[idx - 1].spacegroup, train.iloc[idx - 1].gamma)\n",
    "\n",
    "G = get_crytal_graph(pos_red, pos.element, dists, factor=cg_factor)\n",
    "\n",
    "print \"Spacegroup:\", train.iloc[idx - 1].spacegroup\n",
    "print \"Gamma:\", train.iloc[idx - 1].gamma\n",
    "print \"CG Factor:\", cg_factor\n",
    "print \"Node count:\", G.number_of_nodes()\n",
    "print \"Edge count:\", G.number_of_edges()\n",
    "\n",
    "for i in range(len(pos.element)):\n",
    "    sym_i = pos.element[i]\n",
    "    node_i = '{}_{}'.format(sym_i, i)\n",
    "    crdn_i = list(G.neighbors(node_i))\n",
    "    print \"{:8s}{:d}   {}\".format(node_i, len(crdn_i), crdn_i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'atomic_mass': 69.723, 'cn': 4, 'electroneg': 1.81, 'symbol': 'Ga'}"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "G.node['Ga_0']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'bond_length': 1.808885609695628, 'symbol': 'Ga-O'}"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "G.edges[('Ga_0', 'O_38', 0)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AdjacencyView({'O_40': {0: {'bond_length': 1.8271305422232218, 'symbol': 'Ga-O'}}, 'O_38': {0: {'bond_length': 1.808885609695628, 'symbol': 'Ga-O'}}, 'O_57': {0: {'bond_length': 1.781654092682299, 'symbol': 'Ga-O'}}, 'O_62': {0: {'bond_length': 1.859283310839619, 'symbol': 'Ga-O'}}})"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "G['Ga_0']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ce630d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 10))\n",
    "nx.draw_spring(G, with_labels=True, node_size=800, font_size=8, font_color='w', font_weight='bold');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Get features from CG using Node2Vec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "from gensim.models import Word2Vec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "def learn_embeddings(walks, dimensions=8, workers=4, window=6, iter_num=200):\n",
    "    \"\"\"\n",
    "    Learn embeddings by optimizing the Skipgram objective using SGD.\n",
    "    \"\"\"\n",
    "    walks = [map(str, walk) for walk in walks]\n",
    "    model = Word2Vec(walks, size=dimensions, window=window,\n",
    "                     min_count=0, sg=1, workers=workers, iter=iter_num)\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "G_nv = GraphNV(G, 10, 0.6, is_directed=False, weight_key='bond_length')\n",
    "G_nv.preprocess_transition_probs()\n",
    "walks = G_nv.simulate_walks(10, 4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "result = learn_embeddings(walks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "word_vectors = result.wv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('Ga_10', 0.9236802458763123),\n",
       " ('O_76', 0.8877555131912231),\n",
       " ('O_50', 0.8013238906860352),\n",
       " ('Al_2', 0.7980443835258484),\n",
       " ('O_52', 0.7788082361221313),\n",
       " ('Ga_18', 0.7731853723526001),\n",
       " ('O_60', 0.7121289968490601),\n",
       " ('Al_12', 0.6984312534332275),\n",
       " ('O_59', 0.6962066292762756),\n",
       " ('O_35', 0.685808539390564)]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "word_vectors.most_similar('O_36')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.35862446, -1.159631  ,  1.6005145 ,  1.2689173 , -0.03146712,\n",
       "       -0.9989846 , -0.7158729 ,  1.2908448 ], dtype=float32)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "word_vectors['O_36']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Get features from CG"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "a_list = []\n",
    "for n in G.nodes(data=True):\n",
    "    n[1]['id'] = n[0]\n",
    "    \n",
    "    bls = np.array([G.edges[(n[0], m, 0)]['bond_length'] for m in G.neighbors(n[0])])\n",
    "    n[1]['avg_m-o_bond_length'] = bls.mean()\n",
    "    n[1]['std_m-o_bond_length'] = bls.std()\n",
    "\n",
    "    for i in range(6):\n",
    "        n[1]['v_{}'.format(i)] = word_vectors[n[0]][i]\n",
    "        \n",
    "    a_list.append(n[1])\n",
    "\n",
    "a_df = pd.DataFrame(a_list).set_index('id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "a_df_sum = a_df.groupby('symbol').mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>avg_m-o_bond_length</th>\n",
       "      <th>cn</th>\n",
       "      <th>std_m-o_bond_length</th>\n",
       "      <th>v_0</th>\n",
       "      <th>v_1</th>\n",
       "      <th>v_2</th>\n",
       "      <th>v_3</th>\n",
       "      <th>v_4</th>\n",
       "      <th>v_5</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>symbol</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Al</th>\n",
       "      <td>1.949009</td>\n",
       "      <td>5.500000</td>\n",
       "      <td>0.109757</td>\n",
       "      <td>0.039181</td>\n",
       "      <td>-0.797368</td>\n",
       "      <td>-0.172135</td>\n",
       "      <td>0.660290</td>\n",
       "      <td>-1.004109</td>\n",
       "      <td>-0.639559</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ga</th>\n",
       "      <td>1.943034</td>\n",
       "      <td>5.500000</td>\n",
       "      <td>0.093294</td>\n",
       "      <td>0.061092</td>\n",
       "      <td>-0.981394</td>\n",
       "      <td>0.172217</td>\n",
       "      <td>0.625236</td>\n",
       "      <td>-0.992239</td>\n",
       "      <td>-0.475259</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>O</th>\n",
       "      <td>1.937629</td>\n",
       "      <td>3.666667</td>\n",
       "      <td>0.079646</td>\n",
       "      <td>-0.091446</td>\n",
       "      <td>-0.599316</td>\n",
       "      <td>0.015715</td>\n",
       "      <td>0.886626</td>\n",
       "      <td>-1.085931</td>\n",
       "      <td>-0.597228</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        avg_m-o_bond_length        cn  std_m-o_bond_length       v_0  \\\n",
       "symbol                                                                 \n",
       "Al                 1.949009  5.500000             0.109757  0.039181   \n",
       "Ga                 1.943034  5.500000             0.093294  0.061092   \n",
       "O                  1.937629  3.666667             0.079646 -0.091446   \n",
       "\n",
       "             v_1       v_2       v_3       v_4       v_5  \n",
       "symbol                                                    \n",
       "Al     -0.797368 -0.172135  0.660290 -1.004109 -0.639559  \n",
       "Ga     -0.981394  0.172217  0.625236 -0.992239 -0.475259  \n",
       "O      -0.599316  0.015715  0.886626 -1.085931 -0.597228  "
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a_df_sum.drop(['atomic_mass', 'electroneg'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Process all files (test and train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "def process_graph(batch='train'):\n",
    "    \"\"\"\n",
    "    Function to process each .xyz file and return the crystal graph data\n",
    "    \"\"\"\n",
    "    summary_result = []\n",
    "    \n",
    "    pbar = tqdm_notebook(os.listdir(os.path.join(DATA_DIR, batch)))\n",
    "    \n",
    "    for i in pbar:\n",
    "        # Load .xyz data\n",
    "        fname = os.path.join(DATA_DIR, batch, i, 'geometry.xyz')\n",
    "        pos, latvec, natoms = get_xyz_data(fname)\n",
    "        \n",
    "        # Element index\n",
    "        idx = int(i)\n",
    "        \n",
    "        # Calculate reduced coordinates\n",
    "        pos_R = pos[['x', 'y', 'z']].as_matrix()\n",
    "        pos_red = convert_to_red(pos_R, lat)\n",
    "        \n",
    "        # Get distances between pairs\n",
    "        dists, Rij_min = get_shortest_distances(pos_red, lat)\n",
    "        \n",
    "        # Make CG\n",
    "        if batch == 'train':\n",
    "            d = train.iloc[idx - 1]\n",
    "        elif batch == 'test':\n",
    "            d = test.iloc[idx - 1]\n",
    "        cg_factor = get_factor(d.spacegroup, d.gamma)\n",
    "        G = get_crytal_graph(pos_red, pos.element, dists, factor=cg_factor)\n",
    "        \n",
    "        # Perform N2V with bond_lengths as weights\n",
    "        G_nv = GraphNV(G, 10, 0.6, is_directed=False, weight_key='bond_length')\n",
    "        G_nv.preprocess_transition_probs()\n",
    "        walks = G_nv.simulate_walks(10, 4)\n",
    "        \n",
    "        result = learn_embeddings(walks)\n",
    "        word_vectors = result.wv\n",
    "        \n",
    "        # Create features\n",
    "        a_list = []\n",
    "        for n in G.nodes(data=True):\n",
    "            n[1]['id'] = n[0]\n",
    "\n",
    "            bls = np.array([G.edges[(n[0], m, 0)]['bond_length'] for m in G.neighbors(n[0])])\n",
    "            \n",
    "            if bls.size > 0:\n",
    "                n[1]['avg_m-o_bond_length'] = bls.mean()\n",
    "                n[1]['std_m-o_bond_length'] = bls.std()\n",
    "            else:\n",
    "                n[1]['avg_m-o_bond_length'] = 0.0\n",
    "                n[1]['std_m-o_bond_length'] = 0.0\n",
    "\n",
    "            for k in range(6):\n",
    "                n[1]['v_{}'.format(k)] = word_vectors[n[0]][k]\n",
    "\n",
    "            a_list.append(n[1])\n",
    "        \n",
    "        G.clear()\n",
    "\n",
    "        a_df = pd.DataFrame(a_list).set_index('id')\n",
    "        a_df_sum = a_df.groupby('symbol').mean()\n",
    "        \n",
    "        features = [idx]\n",
    "        for s in ['Al', 'In', 'Ga', 'O']:\n",
    "            try:\n",
    "                features += a_df_sum.drop(['atomic_mass', 'electroneg'], axis=1).loc[s].tolist()\n",
    "            except KeyError:\n",
    "                features += [0.0] * 9\n",
    "        \n",
    "        summary_result.append(features)\n",
    "    \n",
    "    return summary_result"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Process files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9a6f1c2da4fd4bc588d2d0d7cf909f36",
       "version_major": 2,
       "version_minor": 0
      },
      "text/html": [
       "<p>Failed to display Jupyter Widget of type <code>HBox</code>.</p>\n",
       "<p>\n",
       "  If you're reading this message in the Jupyter Notebook or JupyterLab Notebook, it may mean\n",
       "  that the widgets JavaScript is still loading. If this message persists, it\n",
       "  likely means that the widgets JavaScript library is either not installed or\n",
       "  not enabled. See the <a href=\"https://ipywidgets.readthedocs.io/en/stable/user_install.html\">Jupyter\n",
       "  Widgets Documentation</a> for setup instructions.\n",
       "</p>\n",
       "<p>\n",
       "  If you're reading this message in another frontend (for example, a static\n",
       "  rendering on GitHub or <a href=\"https://nbviewer.jupyter.org/\">NBViewer</a>),\n",
       "  it may mean that your frontend doesn't currently support widgets.\n",
       "</p>\n"
      ],
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=2400), HTML(value=u'')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "data_train = process_graph(batch='train')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "63a32bc3288a4c5e94ba7bc9b797a7b1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/html": [
       "<p>Failed to display Jupyter Widget of type <code>HBox</code>.</p>\n",
       "<p>\n",
       "  If you're reading this message in the Jupyter Notebook or JupyterLab Notebook, it may mean\n",
       "  that the widgets JavaScript is still loading. If this message persists, it\n",
       "  likely means that the widgets JavaScript library is either not installed or\n",
       "  not enabled. See the <a href=\"https://ipywidgets.readthedocs.io/en/stable/user_install.html\">Jupyter\n",
       "  Widgets Documentation</a> for setup instructions.\n",
       "</p>\n",
       "<p>\n",
       "  If you're reading this message in another frontend (for example, a static\n",
       "  rendering on GitHub or <a href=\"https://nbviewer.jupyter.org/\">NBViewer</a>),\n",
       "  it may mean that your frontend doesn't currently support widgets.\n",
       "</p>\n"
      ],
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=600), HTML(value=u'')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "data_test = process_graph(batch='test')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.06900000e+03,  1.89209799e+00,  7.80000000e+00, ...,\n",
       "         9.39420525e-01, -7.35231910e-01, -4.19473680e-02],\n",
       "       [ 1.85500000e+03,  1.93980178e+00,  5.44000000e+00, ...,\n",
       "         5.89523919e-01, -9.85791058e-01, -4.05812978e-01],\n",
       "       [ 7.97000000e+02,  2.00616283e+00,  6.00000000e+00, ...,\n",
       "         4.77233331e-01, -2.03986704e-03,  3.22256540e-01],\n",
       "       ...,\n",
       "       [ 1.47600000e+03,  0.00000000e+00,  0.00000000e+00, ...,\n",
       "         1.76482436e-01, -1.24055775e-01,  3.93376606e-01],\n",
       "       [ 1.64400000e+03,  1.96615049e+00,  6.00000000e+00, ...,\n",
       "         1.15421667e+00, -7.65470547e-01, -3.71078826e-01],\n",
       "       [ 9.86000000e+02,  1.82994502e+00,  7.15384615e+00, ...,\n",
       "         9.28527352e-01, -1.06221182e+00, -4.82650069e-01]])"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(data_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "columns = ['id', 'Al_avg_m-o_bond_length', 'Al_avg_cn', 'Al_std_m-o_bond_length',\n",
    "          'Al_v_0', 'Al_v_1', 'Al_v_2', 'Al_v_3', 'Al_v_4', 'Al_v_5',\n",
    "           \n",
    "          'In_avg_m-o_bond_length', 'In_avg_cn', 'In_std_m-o_bond_length',\n",
    "          'In_v_0', 'In_v_1', 'In_v_2', 'In_v_3', 'In_v_4', 'In_v_5',\n",
    "           \n",
    "          'Ga_avg_m-o_bond_length', 'Ga_avg_cn', 'Ga_std_m-o_bond_length',\n",
    "          'Ga_v_0', 'Ga_v_1', 'Ga_v_2', 'Ga_v_3', 'Ga_v_4', 'Ga_v_5',\n",
    "           \n",
    "          'O_avg_m-o_bond_length', 'O_avg_cn', 'O_std_m-o_bond_length',\n",
    "          'O_v_0', 'O_v_1', 'O_v_2', 'O_v_3', 'O_v_4', 'O_v_5',]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train = pd.DataFrame(data_train, columns=columns).sort_values('id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>Al_avg_m-o_bond_length</th>\n",
       "      <th>Al_avg_cn</th>\n",
       "      <th>Al_std_m-o_bond_length</th>\n",
       "      <th>Al_v_0</th>\n",
       "      <th>Al_v_1</th>\n",
       "      <th>Al_v_2</th>\n",
       "      <th>Al_v_3</th>\n",
       "      <th>Al_v_4</th>\n",
       "      <th>Al_v_5</th>\n",
       "      <th>...</th>\n",
       "      <th>Ga_v_5</th>\n",
       "      <th>O_avg_m-o_bond_length</th>\n",
       "      <th>O_avg_cn</th>\n",
       "      <th>O_std_m-o_bond_length</th>\n",
       "      <th>O_v_0</th>\n",
       "      <th>O_v_1</th>\n",
       "      <th>O_v_2</th>\n",
       "      <th>O_v_3</th>\n",
       "      <th>O_v_4</th>\n",
       "      <th>O_v_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1067</th>\n",
       "      <td>1</td>\n",
       "      <td>1.949009</td>\n",
       "      <td>5.500000</td>\n",
       "      <td>0.109757</td>\n",
       "      <td>0.263189</td>\n",
       "      <td>-0.884261</td>\n",
       "      <td>-0.036170</td>\n",
       "      <td>0.750605</td>\n",
       "      <td>-0.698487</td>\n",
       "      <td>-0.661276</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.780323</td>\n",
       "      <td>1.937629</td>\n",
       "      <td>3.666667</td>\n",
       "      <td>0.079646</td>\n",
       "      <td>0.127069</td>\n",
       "      <td>-0.853100</td>\n",
       "      <td>0.163005</td>\n",
       "      <td>0.763118</td>\n",
       "      <td>-0.803645</td>\n",
       "      <td>-0.732734</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2327</th>\n",
       "      <td>2</td>\n",
       "      <td>1.779137</td>\n",
       "      <td>6.500000</td>\n",
       "      <td>0.649896</td>\n",
       "      <td>0.032869</td>\n",
       "      <td>-0.554428</td>\n",
       "      <td>0.408196</td>\n",
       "      <td>1.157920</td>\n",
       "      <td>-1.205161</td>\n",
       "      <td>0.031592</td>\n",
       "      <td>...</td>\n",
       "      <td>0.283309</td>\n",
       "      <td>1.701934</td>\n",
       "      <td>4.687500</td>\n",
       "      <td>0.470738</td>\n",
       "      <td>-0.361125</td>\n",
       "      <td>-0.369090</td>\n",
       "      <td>-0.136622</td>\n",
       "      <td>1.129468</td>\n",
       "      <td>-1.253904</td>\n",
       "      <td>-0.090321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2219</th>\n",
       "      <td>3</td>\n",
       "      <td>1.883655</td>\n",
       "      <td>5.384615</td>\n",
       "      <td>0.224221</td>\n",
       "      <td>-0.312495</td>\n",
       "      <td>-1.304286</td>\n",
       "      <td>0.072648</td>\n",
       "      <td>0.015579</td>\n",
       "      <td>-0.544383</td>\n",
       "      <td>-0.327499</td>\n",
       "      <td>...</td>\n",
       "      <td>0.235588</td>\n",
       "      <td>1.886253</td>\n",
       "      <td>3.666667</td>\n",
       "      <td>0.206266</td>\n",
       "      <td>-0.161973</td>\n",
       "      <td>-1.233899</td>\n",
       "      <td>0.291731</td>\n",
       "      <td>0.097785</td>\n",
       "      <td>-0.647818</td>\n",
       "      <td>-0.186288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2201</th>\n",
       "      <td>4</td>\n",
       "      <td>2.012969</td>\n",
       "      <td>5.555556</td>\n",
       "      <td>0.141713</td>\n",
       "      <td>-0.371033</td>\n",
       "      <td>-0.838455</td>\n",
       "      <td>0.283264</td>\n",
       "      <td>0.640824</td>\n",
       "      <td>-0.380119</td>\n",
       "      <td>-0.257924</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.023984</td>\n",
       "      <td>3.611111</td>\n",
       "      <td>0.115877</td>\n",
       "      <td>-0.578469</td>\n",
       "      <td>-0.738934</td>\n",
       "      <td>0.272375</td>\n",
       "      <td>0.632710</td>\n",
       "      <td>-0.161818</td>\n",
       "      <td>-0.045243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2342</th>\n",
       "      <td>5</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.524899</td>\n",
       "      <td>1.813465</td>\n",
       "      <td>4.958333</td>\n",
       "      <td>0.536679</td>\n",
       "      <td>-0.177194</td>\n",
       "      <td>-0.745338</td>\n",
       "      <td>0.727849</td>\n",
       "      <td>0.925179</td>\n",
       "      <td>-0.707048</td>\n",
       "      <td>-0.737106</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      id  Al_avg_m-o_bond_length  Al_avg_cn  Al_std_m-o_bond_length    Al_v_0  \\\n",
       "1067   1                1.949009   5.500000                0.109757  0.263189   \n",
       "2327   2                1.779137   6.500000                0.649896  0.032869   \n",
       "2219   3                1.883655   5.384615                0.224221 -0.312495   \n",
       "2201   4                2.012969   5.555556                0.141713 -0.371033   \n",
       "2342   5                0.000000   0.000000                0.000000  0.000000   \n",
       "\n",
       "        Al_v_1    Al_v_2    Al_v_3    Al_v_4    Al_v_5    ...       Ga_v_5  \\\n",
       "1067 -0.884261 -0.036170  0.750605 -0.698487 -0.661276    ...    -0.780323   \n",
       "2327 -0.554428  0.408196  1.157920 -1.205161  0.031592    ...     0.283309   \n",
       "2219 -1.304286  0.072648  0.015579 -0.544383 -0.327499    ...     0.235588   \n",
       "2201 -0.838455  0.283264  0.640824 -0.380119 -0.257924    ...     0.000000   \n",
       "2342  0.000000  0.000000  0.000000  0.000000  0.000000    ...    -0.524899   \n",
       "\n",
       "      O_avg_m-o_bond_length  O_avg_cn  O_std_m-o_bond_length     O_v_0  \\\n",
       "1067               1.937629  3.666667               0.079646  0.127069   \n",
       "2327               1.701934  4.687500               0.470738 -0.361125   \n",
       "2219               1.886253  3.666667               0.206266 -0.161973   \n",
       "2201               2.023984  3.611111               0.115877 -0.578469   \n",
       "2342               1.813465  4.958333               0.536679 -0.177194   \n",
       "\n",
       "         O_v_1     O_v_2     O_v_3     O_v_4     O_v_5  \n",
       "1067 -0.853100  0.163005  0.763118 -0.803645 -0.732734  \n",
       "2327 -0.369090 -0.136622  1.129468 -1.253904 -0.090321  \n",
       "2219 -1.233899  0.291731  0.097785 -0.647818 -0.186288  \n",
       "2201 -0.738934  0.272375  0.632710 -0.161818 -0.045243  \n",
       "2342 -0.745338  0.727849  0.925179 -0.707048 -0.737106  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_test = pd.DataFrame(data_test, columns=columns).sort_values('id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>Al_avg_m-o_bond_length</th>\n",
       "      <th>Al_avg_cn</th>\n",
       "      <th>Al_std_m-o_bond_length</th>\n",
       "      <th>Al_v_0</th>\n",
       "      <th>Al_v_1</th>\n",
       "      <th>Al_v_2</th>\n",
       "      <th>Al_v_3</th>\n",
       "      <th>Al_v_4</th>\n",
       "      <th>Al_v_5</th>\n",
       "      <th>...</th>\n",
       "      <th>Ga_v_5</th>\n",
       "      <th>O_avg_m-o_bond_length</th>\n",
       "      <th>O_avg_cn</th>\n",
       "      <th>O_std_m-o_bond_length</th>\n",
       "      <th>O_v_0</th>\n",
       "      <th>O_v_1</th>\n",
       "      <th>O_v_2</th>\n",
       "      <th>O_v_3</th>\n",
       "      <th>O_v_4</th>\n",
       "      <th>O_v_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>260</th>\n",
       "      <td>1</td>\n",
       "      <td>1.997013</td>\n",
       "      <td>5.666667</td>\n",
       "      <td>0.149430</td>\n",
       "      <td>-0.229542</td>\n",
       "      <td>-0.690060</td>\n",
       "      <td>-0.071598</td>\n",
       "      <td>1.570191</td>\n",
       "      <td>-0.040740</td>\n",
       "      <td>-0.687938</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.722323</td>\n",
       "      <td>2.039899</td>\n",
       "      <td>4.083333</td>\n",
       "      <td>0.259720</td>\n",
       "      <td>0.060481</td>\n",
       "      <td>-0.246262</td>\n",
       "      <td>0.042607</td>\n",
       "      <td>1.223458</td>\n",
       "      <td>-0.488761</td>\n",
       "      <td>-0.947568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>578</th>\n",
       "      <td>2</td>\n",
       "      <td>1.946632</td>\n",
       "      <td>5.500000</td>\n",
       "      <td>0.104698</td>\n",
       "      <td>0.240536</td>\n",
       "      <td>-0.306924</td>\n",
       "      <td>0.042656</td>\n",
       "      <td>0.527508</td>\n",
       "      <td>-1.204234</td>\n",
       "      <td>-0.554868</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.607407</td>\n",
       "      <td>1.937739</td>\n",
       "      <td>3.666667</td>\n",
       "      <td>0.077864</td>\n",
       "      <td>0.218927</td>\n",
       "      <td>-0.319189</td>\n",
       "      <td>0.078719</td>\n",
       "      <td>0.532809</td>\n",
       "      <td>-1.276648</td>\n",
       "      <td>-0.548699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>543</th>\n",
       "      <td>3</td>\n",
       "      <td>2.016860</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>0.125847</td>\n",
       "      <td>-0.401368</td>\n",
       "      <td>-0.606386</td>\n",
       "      <td>0.310870</td>\n",
       "      <td>0.574860</td>\n",
       "      <td>-0.258066</td>\n",
       "      <td>0.559656</td>\n",
       "      <td>...</td>\n",
       "      <td>0.647948</td>\n",
       "      <td>2.016307</td>\n",
       "      <td>3.611111</td>\n",
       "      <td>0.102570</td>\n",
       "      <td>-0.635103</td>\n",
       "      <td>-0.748690</td>\n",
       "      <td>0.325465</td>\n",
       "      <td>0.460211</td>\n",
       "      <td>-0.197821</td>\n",
       "      <td>0.350782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>539</th>\n",
       "      <td>4</td>\n",
       "      <td>1.851650</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>0.391329</td>\n",
       "      <td>-0.253058</td>\n",
       "      <td>-0.822644</td>\n",
       "      <td>0.014520</td>\n",
       "      <td>0.620034</td>\n",
       "      <td>-1.219882</td>\n",
       "      <td>-0.182529</td>\n",
       "      <td>...</td>\n",
       "      <td>0.144882</td>\n",
       "      <td>1.837372</td>\n",
       "      <td>5.125000</td>\n",
       "      <td>0.467283</td>\n",
       "      <td>-0.000689</td>\n",
       "      <td>-0.711740</td>\n",
       "      <td>0.227946</td>\n",
       "      <td>0.971360</td>\n",
       "      <td>-1.139314</td>\n",
       "      <td>-0.059066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>583</th>\n",
       "      <td>5</td>\n",
       "      <td>1.841404</td>\n",
       "      <td>6.333333</td>\n",
       "      <td>0.363869</td>\n",
       "      <td>-0.578203</td>\n",
       "      <td>-1.231069</td>\n",
       "      <td>0.219664</td>\n",
       "      <td>0.778618</td>\n",
       "      <td>-0.456018</td>\n",
       "      <td>-0.183017</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.069742</td>\n",
       "      <td>1.942138</td>\n",
       "      <td>5.187500</td>\n",
       "      <td>0.484302</td>\n",
       "      <td>-0.489022</td>\n",
       "      <td>-1.321007</td>\n",
       "      <td>0.086943</td>\n",
       "      <td>0.762741</td>\n",
       "      <td>-0.402170</td>\n",
       "      <td>-0.122682</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     id  Al_avg_m-o_bond_length  Al_avg_cn  Al_std_m-o_bond_length    Al_v_0  \\\n",
       "260   1                1.997013   5.666667                0.149430 -0.229542   \n",
       "578   2                1.946632   5.500000                0.104698  0.240536   \n",
       "543   3                2.016860   6.000000                0.125847 -0.401368   \n",
       "539   4                1.851650   7.000000                0.391329 -0.253058   \n",
       "583   5                1.841404   6.333333                0.363869 -0.578203   \n",
       "\n",
       "       Al_v_1    Al_v_2    Al_v_3    Al_v_4    Al_v_5    ...       Ga_v_5  \\\n",
       "260 -0.690060 -0.071598  1.570191 -0.040740 -0.687938    ...    -0.722323   \n",
       "578 -0.306924  0.042656  0.527508 -1.204234 -0.554868    ...    -0.607407   \n",
       "543 -0.606386  0.310870  0.574860 -0.258066  0.559656    ...     0.647948   \n",
       "539 -0.822644  0.014520  0.620034 -1.219882 -0.182529    ...     0.144882   \n",
       "583 -1.231069  0.219664  0.778618 -0.456018 -0.183017    ...    -0.069742   \n",
       "\n",
       "     O_avg_m-o_bond_length  O_avg_cn  O_std_m-o_bond_length     O_v_0  \\\n",
       "260               2.039899  4.083333               0.259720  0.060481   \n",
       "578               1.937739  3.666667               0.077864  0.218927   \n",
       "543               2.016307  3.611111               0.102570 -0.635103   \n",
       "539               1.837372  5.125000               0.467283 -0.000689   \n",
       "583               1.942138  5.187500               0.484302 -0.489022   \n",
       "\n",
       "        O_v_1     O_v_2     O_v_3     O_v_4     O_v_5  \n",
       "260 -0.246262  0.042607  1.223458 -0.488761 -0.947568  \n",
       "578 -0.319189  0.078719  0.532809 -1.276648 -0.548699  \n",
       "543 -0.748690  0.325465  0.460211 -0.197821  0.350782  \n",
       "539 -0.711740  0.227946  0.971360 -1.139314 -0.059066  \n",
       "583 -1.321007  0.086943  0.762741 -0.402170 -0.122682  \n",
       "\n",
       "[5 rows x 37 columns]"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Write out geo test and train data files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train.to_csv(os.path.join(DATA_DIR, 'train_geo.csv'), index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_test.to_csv(os.path.join(DATA_DIR, 'test_geo.csv'), index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}