pymatgen_matminer.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "708a4288",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pymatgen.ext.matproj import MPRester\n",
    "from pymatgen.io.cif import CifParser\n",
    "import matminer\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%config InlineBackend.figure_format='retina'\n",
    "%matplotlib Inline"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5eeac254",
   "metadata": {},
   "source": [
    "# Pymatgen"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ab9629aa",
   "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>pretty_formula</th>\n",
       "      <th>formation_energy_per_atom</th>\n",
       "      <th>band_gap</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KO2</td>\n",
       "      <td>-0.950213</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KO3</td>\n",
       "      <td>-0.656296</td>\n",
       "      <td>0.7121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>K2O</td>\n",
       "      <td>-1.193563</td>\n",
       "      <td>1.8534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>K2O</td>\n",
       "      <td>-1.213501</td>\n",
       "      <td>0.4029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>KO2</td>\n",
       "      <td>-0.926555</td>\n",
       "      <td>0.1546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>NaO3</td>\n",
       "      <td>-0.603353</td>\n",
       "      <td>0.6243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>NaO2</td>\n",
       "      <td>-0.884666</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>Na2O2</td>\n",
       "      <td>-1.078611</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>NaO2</td>\n",
       "      <td>-0.830114</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62</th>\n",
       "      <td>Na2O</td>\n",
       "      <td>-1.371527</td>\n",
       "      <td>1.1491</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>63 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   pretty_formula  formation_energy_per_atom  band_gap\n",
       "0             KO2                  -0.950213    0.0000\n",
       "1             KO3                  -0.656296    0.7121\n",
       "2             K2O                  -1.193563    1.8534\n",
       "3             K2O                  -1.213501    0.4029\n",
       "4             KO2                  -0.926555    0.1546\n",
       "..            ...                        ...       ...\n",
       "58           NaO3                  -0.603353    0.6243\n",
       "59           NaO2                  -0.884666    0.0000\n",
       "60          Na2O2                  -1.078611    0.0000\n",
       "61           NaO2                  -0.830114    0.0000\n",
       "62           Na2O                  -1.371527    1.1491\n",
       "\n",
       "[63 rows x 3 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with MPRester(\"....\") as mpr:\n",
    "    criteria = {\"elements\":{\"$in\":[\"Li\", \"Na\", \"K\"], \"$all\": [\"O\"]}, \"nelements\":2}\n",
    "    properties = [\"pretty_formula\", \"formation_energy_per_atom\", \"band_gap\"]\n",
    "    shortC = (mpr.query(criteria, properties))\n",
    "    \n",
    "data = pd.DataFrame(shortC)\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9addb27e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 261,
       "width": 376
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(data[\"formation_energy_per_atom\"], data[\"band_gap\"], 'ro')\n",
    "plt.xlabel(\"Formation energy (eV)\")\n",
    "plt.ylabel(\"Band gap (eV)\")\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "efaee061",
   "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>task_id</th>\n",
       "      <th>spacegroup.symbol</th>\n",
       "      <th>final_energy</th>\n",
       "      <th>band_gap</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-558125</td>\n",
       "      <td>C222_1</td>\n",
       "      <td>-240.251746</td>\n",
       "      <td>2.0811</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-5777</td>\n",
       "      <td>Amm2</td>\n",
       "      <td>-40.056169</td>\n",
       "      <td>2.3733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-2998</td>\n",
       "      <td>Pm-3m</td>\n",
       "      <td>-39.984795</td>\n",
       "      <td>1.7252</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-1076932</td>\n",
       "      <td>Amm2</td>\n",
       "      <td>-35.163458</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-504715</td>\n",
       "      <td>Pm-3m</td>\n",
       "      <td>-24.641710</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>mp-995191</td>\n",
       "      <td>Amm2</td>\n",
       "      <td>-31.025462</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>mp-19990</td>\n",
       "      <td>P4/mmm</td>\n",
       "      <td>-31.021441</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>mp-644497</td>\n",
       "      <td>Amm2</td>\n",
       "      <td>-36.487377</td>\n",
       "      <td>0.5466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>mp-5933</td>\n",
       "      <td>P6_3/mmc</td>\n",
       "      <td>-240.099121</td>\n",
       "      <td>1.7261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>mp-5020</td>\n",
       "      <td>R3m</td>\n",
       "      <td>-40.058535</td>\n",
       "      <td>2.6212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>mp-5986</td>\n",
       "      <td>P4mm</td>\n",
       "      <td>-40.042498</td>\n",
       "      <td>1.8298</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       task_id spacegroup.symbol  final_energy  band_gap\n",
       "0    mp-558125            C222_1   -240.251746    2.0811\n",
       "1      mp-5777              Amm2    -40.056169    2.3733\n",
       "2      mp-2998             Pm-3m    -39.984795    1.7252\n",
       "3   mp-1076932              Amm2    -35.163458    0.0000\n",
       "4    mp-504715             Pm-3m    -24.641710    0.0000\n",
       "5    mp-995191              Amm2    -31.025462    0.0000\n",
       "6     mp-19990            P4/mmm    -31.021441    0.0000\n",
       "7    mp-644497              Amm2    -36.487377    0.5466\n",
       "8      mp-5933          P6_3/mmc   -240.099121    1.7261\n",
       "9      mp-5020               R3m    -40.058535    2.6212\n",
       "10     mp-5986              P4mm    -40.042498    1.8298"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with MPRester(\"....\") as mpr:\n",
    "    criteria = {\"pretty_formula\": \"BaTiO3\"}\n",
    "    properties = [\"task_id\", \"spacegroup.symbol\", \"final_energy\", \"band_gap\"]\n",
    "    shortC = (mpr.query(criteria, properties))\n",
    "    \n",
    "data = pd.DataFrame(shortC)\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "826a0f79",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 372
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(data[\"band_gap\"])\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12940f33",
   "metadata": {},
   "source": [
    "## Oxide Perovskites\n",
    "\n",
    "$ABO_3$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "60b1c718",
   "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>energy</th>\n",
       "      <th>energy_per_atom</th>\n",
       "      <th>volume</th>\n",
       "      <th>formation_energy_per_atom</th>\n",
       "      <th>nsites</th>\n",
       "      <th>unit_cell_formula</th>\n",
       "      <th>pretty_formula</th>\n",
       "      <th>is_hubbard</th>\n",
       "      <th>elements</th>\n",
       "      <th>nelements</th>\n",
       "      <th>...</th>\n",
       "      <th>cif</th>\n",
       "      <th>total_magnetization</th>\n",
       "      <th>material_id</th>\n",
       "      <th>oxide_type</th>\n",
       "      <th>tags</th>\n",
       "      <th>elasticity</th>\n",
       "      <th>piezo</th>\n",
       "      <th>diel</th>\n",
       "      <th>deprecated</th>\n",
       "      <th>full_formula</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-39.099719</td>\n",
       "      <td>-7.819944</td>\n",
       "      <td>58.671145</td>\n",
       "      <td>-3.699194</td>\n",
       "      <td>5</td>\n",
       "      <td>{'Ac': 1.0, 'Al': 1.0, 'O': 3.0}</td>\n",
       "      <td>AcAlO3</td>\n",
       "      <td>False</td>\n",
       "      <td>[Ac, Al, O]</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td># generated using pymatgen\\ndata_AcAlO3\\n_symm...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>mp-1183115</td>\n",
       "      <td>oxide</td>\n",
       "      <td>[]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>Ac1Al1O3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-35.960398</td>\n",
       "      <td>-7.192080</td>\n",
       "      <td>51.548126</td>\n",
       "      <td>-2.484567</td>\n",
       "      <td>5</td>\n",
       "      <td>{'Ac': 1.0, 'B': 1.0, 'O': 3.0}</td>\n",
       "      <td>AcBO3</td>\n",
       "      <td>False</td>\n",
       "      <td>[Ac, B, O]</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td># generated using pymatgen\\ndata_AcBO3\\n_symme...</td>\n",
       "      <td>0.000133</td>\n",
       "      <td>mp-1183052</td>\n",
       "      <td>oxide</td>\n",
       "      <td>[]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>Ac1B1O3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-40.252962</td>\n",
       "      <td>-8.050592</td>\n",
       "      <td>63.488393</td>\n",
       "      <td>-3.150949</td>\n",
       "      <td>5</td>\n",
       "      <td>{'Ac': 1.0, 'Cr': 1.0, 'O': 3.0}</td>\n",
       "      <td>AcCrO3</td>\n",
       "      <td>True</td>\n",
       "      <td>[Ac, Cr, O]</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td># generated using pymatgen\\ndata_AcCrO3\\n_symm...</td>\n",
       "      <td>3.000033</td>\n",
       "      <td>mp-866101</td>\n",
       "      <td>oxide</td>\n",
       "      <td>[]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>Ac1Cr1O3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-33.117727</td>\n",
       "      <td>-6.623545</td>\n",
       "      <td>61.365464</td>\n",
       "      <td>-2.432070</td>\n",
       "      <td>5</td>\n",
       "      <td>{'Ac': 1.0, 'Cu': 1.0, 'O': 3.0}</td>\n",
       "      <td>AcCuO3</td>\n",
       "      <td>False</td>\n",
       "      <td>[Ac, Cu, O]</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td># generated using pymatgen\\ndata_AcCuO3\\n_symm...</td>\n",
       "      <td>0.000028</td>\n",
       "      <td>mp-864606</td>\n",
       "      <td>oxide</td>\n",
       "      <td>[]</td>\n",
       "      <td>{'G_Reuss': 84.0, 'G_VRH': 84.0, 'G_Voigt': 85...</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>Ac1Cu1O3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-36.975717</td>\n",
       "      <td>-7.395143</td>\n",
       "      <td>64.691419</td>\n",
       "      <td>-2.876250</td>\n",
       "      <td>5</td>\n",
       "      <td>{'Ac': 1.0, 'Fe': 1.0, 'O': 3.0}</td>\n",
       "      <td>AcFeO3</td>\n",
       "      <td>True</td>\n",
       "      <td>[Ac, Fe, O]</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td># generated using pymatgen\\ndata_AcFeO3\\n_symm...</td>\n",
       "      <td>4.999780</td>\n",
       "      <td>mp-861502</td>\n",
       "      <td>oxide</td>\n",
       "      <td>[]</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>Ac1Fe1O3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      energy  energy_per_atom     volume  formation_energy_per_atom  nsites  \\\n",
       "0 -39.099719        -7.819944  58.671145                  -3.699194       5   \n",
       "1 -35.960398        -7.192080  51.548126                  -2.484567       5   \n",
       "2 -40.252962        -8.050592  63.488393                  -3.150949       5   \n",
       "3 -33.117727        -6.623545  61.365464                  -2.432070       5   \n",
       "4 -36.975717        -7.395143  64.691419                  -2.876250       5   \n",
       "\n",
       "                  unit_cell_formula pretty_formula  is_hubbard     elements  \\\n",
       "0  {'Ac': 1.0, 'Al': 1.0, 'O': 3.0}         AcAlO3       False  [Ac, Al, O]   \n",
       "1   {'Ac': 1.0, 'B': 1.0, 'O': 3.0}          AcBO3       False   [Ac, B, O]   \n",
       "2  {'Ac': 1.0, 'Cr': 1.0, 'O': 3.0}         AcCrO3        True  [Ac, Cr, O]   \n",
       "3  {'Ac': 1.0, 'Cu': 1.0, 'O': 3.0}         AcCuO3       False  [Ac, Cu, O]   \n",
       "4  {'Ac': 1.0, 'Fe': 1.0, 'O': 3.0}         AcFeO3        True  [Ac, Fe, O]   \n",
       "\n",
       "   nelements  ...                                                cif  \\\n",
       "0          3  ...  # generated using pymatgen\\ndata_AcAlO3\\n_symm...   \n",
       "1          3  ...  # generated using pymatgen\\ndata_AcBO3\\n_symme...   \n",
       "2          3  ...  # generated using pymatgen\\ndata_AcCrO3\\n_symm...   \n",
       "3          3  ...  # generated using pymatgen\\ndata_AcCuO3\\n_symm...   \n",
       "4          3  ...  # generated using pymatgen\\ndata_AcFeO3\\n_symm...   \n",
       "\n",
       "  total_magnetization  material_id oxide_type tags  \\\n",
       "0            0.000000   mp-1183115      oxide   []   \n",
       "1            0.000133   mp-1183052      oxide   []   \n",
       "2            3.000033    mp-866101      oxide   []   \n",
       "3            0.000028    mp-864606      oxide   []   \n",
       "4            4.999780    mp-861502      oxide   []   \n",
       "\n",
       "                                          elasticity  piezo  diel deprecated  \\\n",
       "0                                               None   None  None      False   \n",
       "1                                               None   None  None      False   \n",
       "2                                               None   None  None      False   \n",
       "3  {'G_Reuss': 84.0, 'G_VRH': 84.0, 'G_Voigt': 85...   None  None      False   \n",
       "4                                               None   None  None      False   \n",
       "\n",
       "  full_formula  \n",
       "0     Ac1Al1O3  \n",
       "1      Ac1B1O3  \n",
       "2     Ac1Cr1O3  \n",
       "3     Ac1Cu1O3  \n",
       "4     Ac1Fe1O3  \n",
       "\n",
       "[5 rows x 29 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with MPRester(\"....\") as mpr:\n",
    "    perov = mpr.get_data(\"**O3\")\n",
    "    \n",
    "data = pd.DataFrame(perov)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f20170a5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 2398 entries, 0 to 2397\n",
      "Data columns (total 29 columns):\n",
      " #   Column                     Non-Null Count  Dtype  \n",
      "---  ------                     --------------  -----  \n",
      " 0   energy                     2398 non-null   float64\n",
      " 1   energy_per_atom            2398 non-null   float64\n",
      " 2   volume                     2398 non-null   float64\n",
      " 3   formation_energy_per_atom  2398 non-null   float64\n",
      " 4   nsites                     2398 non-null   int64  \n",
      " 5   unit_cell_formula          2398 non-null   object \n",
      " 6   pretty_formula             2398 non-null   object \n",
      " 7   is_hubbard                 2398 non-null   bool   \n",
      " 8   elements                   2398 non-null   object \n",
      " 9   nelements                  2398 non-null   int64  \n",
      " 10  e_above_hull               2398 non-null   float64\n",
      " 11  hubbards                   2398 non-null   object \n",
      " 12  is_compatible              2398 non-null   bool   \n",
      " 13  spacegroup                 2398 non-null   object \n",
      " 14  task_ids                   2398 non-null   object \n",
      " 15  band_gap                   2398 non-null   float64\n",
      " 16  density                    2398 non-null   float64\n",
      " 17  icsd_id                    0 non-null      object \n",
      " 18  icsd_ids                   2398 non-null   object \n",
      " 19  cif                        2398 non-null   object \n",
      " 20  total_magnetization        2398 non-null   float64\n",
      " 21  material_id                2398 non-null   object \n",
      " 22  oxide_type                 2398 non-null   object \n",
      " 23  tags                       2398 non-null   object \n",
      " 24  elasticity                 524 non-null    object \n",
      " 25  piezo                      104 non-null    object \n",
      " 26  diel                       205 non-null    object \n",
      " 27  deprecated                 2398 non-null   bool   \n",
      " 28  full_formula               2398 non-null   object \n",
      "dtypes: bool(3), float64(8), int64(2), object(16)\n",
      "memory usage: 494.2+ KB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ef22e56a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>material_id</th>\n",
       "      <th>pretty_formula</th>\n",
       "      <th>energy_per_atom</th>\n",
       "      <th>band_gap</th>\n",
       "      <th>formation_energy_per_atom</th>\n",
       "      <th>e_above_hull</th>\n",
       "      <th>elements</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-1183115</td>\n",
       "      <td>AcAlO3</td>\n",
       "      <td>-7.819944</td>\n",
       "      <td>4.1024</td>\n",
       "      <td>-3.699194</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>[Ac, Al, O]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-1183052</td>\n",
       "      <td>AcBO3</td>\n",
       "      <td>-7.192080</td>\n",
       "      <td>0.8127</td>\n",
       "      <td>-2.484567</td>\n",
       "      <td>0.792473</td>\n",
       "      <td>[Ac, B, O]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-866101</td>\n",
       "      <td>AcCrO3</td>\n",
       "      <td>-8.050592</td>\n",
       "      <td>2.0832</td>\n",
       "      <td>-3.150949</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>[Ac, Cr, O]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-864606</td>\n",
       "      <td>AcCuO3</td>\n",
       "      <td>-6.623545</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-2.432070</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>[Ac, Cu, O]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-861502</td>\n",
       "      <td>AcFeO3</td>\n",
       "      <td>-7.395143</td>\n",
       "      <td>1.1671</td>\n",
       "      <td>-2.876250</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>[Ac, Fe, O]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id pretty_formula  energy_per_atom  band_gap  \\\n",
       "0  mp-1183115         AcAlO3        -7.819944    4.1024   \n",
       "1  mp-1183052          AcBO3        -7.192080    0.8127   \n",
       "2   mp-866101         AcCrO3        -8.050592    2.0832   \n",
       "3   mp-864606         AcCuO3        -6.623545    0.0000   \n",
       "4   mp-861502         AcFeO3        -7.395143    1.1671   \n",
       "\n",
       "   formation_energy_per_atom  e_above_hull     elements  \n",
       "0                  -3.699194      0.000000  [Ac, Al, O]  \n",
       "1                  -2.484567      0.792473   [Ac, B, O]  \n",
       "2                  -3.150949      0.000000  [Ac, Cr, O]  \n",
       "3                  -2.432070      0.000000  [Ac, Cu, O]  \n",
       "4                  -2.876250      0.000000  [Ac, Fe, O]  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "perovskites = data[[\"material_id\", \"pretty_formula\", \"energy_per_atom\", \"band_gap\", \"formation_energy_per_atom\",\n",
    "                   \"e_above_hull\", \"elements\"]]\n",
    "perovskites.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d1ebe24a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 4 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 357,
       "width": 486
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 6))\n",
    "\n",
    "plt.subplot(221)\n",
    "plt.hist(perovskites[\"energy_per_atom\"], label=\"energy_per_atom\", color=\"red\")\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(222)\n",
    "plt.hist(perovskites[\"band_gap\"], label=\"band_gap\", color=\"blue\")\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(223)\n",
    "plt.hist(perovskites[\"formation_energy_per_atom\"], label=\"formation_energy_per_atom\", color=\"green\")\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(224)\n",
    "plt.hist(perovskites[\"e_above_hull\"], label=\"e_above_hull\", color=\"orange\")\n",
    "plt.legend()\n",
    "\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ede4e00e",
   "metadata": {},
   "source": [
    "### Train ML"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "2f3ba575",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 375
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "y = perovskites[\"band_gap\"].values\n",
    "X = perovskites.drop([\"material_id\", \"pretty_formula\", \"elements\"], axis=1)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=42)\n",
    "\n",
    "\n",
    "plt.hist(y_train, label=\"y_train\")\n",
    "plt.hist(y_test, label=\"y_train\")\n",
    "plt.legend()\n",
    "\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "3e25e663",
   "metadata": {},
   "outputs": [],
   "source": [
    "def algorithm(algo):\n",
    "    y = perovskites[\"band_gap\"].values\n",
    "    X = perovskites.drop([\"material_id\", \"pretty_formula\", \"elements\"], axis=1)\n",
    "\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=42)\n",
    "    \n",
    "    algo =RandomForestRegressor()\n",
    "    algo.fit(X_train, y_train)\n",
    "\n",
    "    y_trained_pred = algo.predict(X_train)\n",
    "    y_test_pred = algo.predict(X_test)\n",
    "\n",
    "    from sklearn.metrics import mean_squared_error\n",
    "\n",
    "    mse_train = mean_squared_error(y_train, y_trained_pred)\n",
    "    mse_test = mean_squared_error(y_test, y_test_pred)\n",
    "\n",
    "    print(\"Training RMSE = {:.5f} eV/atom\".format(np.sqrt(mse_train)))\n",
    "    print(\"Test RMSE = {:.5f} eV/atom\".format(np.sqrt(mse_test)))\n",
    "\n",
    "    plt.plot(y_train*100, y_trained_pred*100, 'ro', label=\"Training set\")\n",
    "    plt.plot(y_test*100, y_test_pred*100, 'bo', label=\"Test set\")\n",
    "    plt.plot([0, 10], [0, 10], 'k-')\n",
    "    plt.title(\"{}\".format(algo))\n",
    "    plt.xlim(0, 6)\n",
    "    plt.ylim(0, 6)\n",
    "    plt.ylabel(\"Predicted (x $10^{-2}$ eV)\")\n",
    "    plt.xlabel(\"Input (x $10^{-2}$ eV)\")\n",
    "    plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "32a5a492",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training RMSE = 0.00224 eV/atom\n",
      "Test RMSE = 0.00468 eV/atom\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 280,
       "width": 381
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "algorithm(RandomForestRegressor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "88e8afdd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training RMSE = 0.00219 eV/atom\n",
      "Test RMSE = 0.00472 eV/atom\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 280,
       "width": 381
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.linear_model import BayesianRidge, ARDRegression, Lasso, LinearRegression, LogisticRegression, ridge_regression\n",
    "\n",
    "algorithm(BayesianRidge)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "9a141966",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training RMSE = 0.00237 eV/atom\n",
      "Test RMSE = 0.00479 eV/atom\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 280,
       "width": 381
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "algorithm(ARDRegression)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "87bc3276",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training RMSE = 0.00227 eV/atom\n",
      "Test RMSE = 0.00450 eV/atom\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 280,
       "width": 381
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "algorithm(Lasso)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "a10fcb79",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training RMSE = 0.00230 eV/atom\n",
      "Test RMSE = 0.00465 eV/atom\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 280,
       "width": 381
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "algorithm(LinearRegression)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "7d07fd8f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training RMSE = 0.00258 eV/atom\n",
      "Test RMSE = 0.00462 eV/atom\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 280,
       "width": 381
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "algorithm(LogisticRegression)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "c93f8348",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training RMSE = 0.00239 eV/atom\n",
      "Test RMSE = 0.00466 eV/atom\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 280,
       "width": 381
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "algorithm(ridge_regression)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c228d6fd",
   "metadata": {},
   "source": [
    "Materials data access Taylor spark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "bfd956dc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "28.0855 amu\n",
      "1.1 ang\n"
     ]
    }
   ],
   "source": [
    "# with MPRester(\"nb3XlTaW98NGIOTg\") as mpr:\n",
    "\n",
    "import pymatgen.core as mg\n",
    "si = mg.Element(\"Si\")\n",
    "print(si.atomic_mass)\n",
    "print(si.atomic_radius)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "0a5d9609",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['mp-1009835', 'mp-1009832', 'mp-1086', 'mp-1009817'] ['mp-611448', 'mp-1097832', 'mp-1205283', 'mp-24', 'mp-1078845', 'mp-1244913', 'mp-1008374', 'mp-579909', 'mp-1181996', 'mp-1194362', 'mp-1188817', 'mp-1182684', 'mp-1197903', 'mp-568286', 'mp-683919', 'mp-1008395', 'mp-569416', 'mp-568806', 'mp-606949', 'mp-611426', 'mp-1147718', 'mp-570002', 'mp-616440', 'mp-1018088', 'mp-1040425', 'mp-937760', 'mp-990448', 'mp-1096869', 'mp-568028', 'mp-66', 'mp-997182', 'mp-1080826', 'mp-1203645', 'mp-1182029', 'mp-998866', 'mp-47', 'mp-630227', 'mp-1205417', 'mp-1244964', 'mp-569517', 'mp-1095534', 'mp-1196583', 'mp-680372', 'mp-169', 'mp-624889', 'mp-1245190', 'mp-667273', 'mp-568363', 'mp-1056957', 'mp-569304', 'mp-568410', 'mp-569567', 'mp-48', 'mp-1095633', 'mp-632329', 'mp-990424', 'mp-1190171', 'mp-1192619', 'mp-1009835', 'mp-1025192', 'mp-1009832', 'mp-1218120', 'mp-1218000', 'mp-1086', 'mp-7088', 'mp-1009817', 'mp-1187218', 'mp-6986', 'mp-569794', 'mp-42', 'mp-1076923', 'mp-50', 'mp-697196', 'mp-1217756', 'mp-1095086']\n"
     ]
    }
   ],
   "source": [
    "from pymatgen.ext.matproj import MPRester\n",
    "\n",
    "with MPRester(\".....\") as mpr:\n",
    "    a = mpr.get_materials_ids(\"TaC\")\n",
    "    b = mpr.get_materials_ids(\"Ta-C\")\n",
    "print(a, b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "7c9993b2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Full Formula (Ta1 C1)\n",
      "Reduced Formula: TaC\n",
      "abc   :   3.169306   3.169306   3.169306\n",
      "angles:  60.000000  60.000000  60.000000\n",
      "Sites (2)\n",
      "  #  SP      a    b    c    magmom\n",
      "---  ----  ---  ---  ---  --------\n",
      "  0  Ta    0    0    0          -0\n",
      "  1  C     0.5  0.5  0.5         0\n",
      "Cohesive Energy = 17.55140979\n"
     ]
    }
   ],
   "source": [
    "with MPRester(\".....\") as mpr:\n",
    "    structure = mpr.get_structure_by_material_id(\"mp-1086\")\n",
    "    cohesive_energy = mpr.get_cohesive_energy(\"mp-1086\")\n",
    "    \n",
    "print(structure)\n",
    "print(\"Cohesive Energy = {}\".format(cohesive_energy))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b388e61a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>material_id</th>\n",
       "      <th>pretty_formula</th>\n",
       "      <th>band_gap</th>\n",
       "      <th>nelements</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-995203</td>\n",
       "      <td>H6PbCI3N</td>\n",
       "      <td>1.7575</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-1194995</td>\n",
       "      <td>H36Pb3C10(I5N2)2</td>\n",
       "      <td>1.7045</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-995214</td>\n",
       "      <td>H6PbCI3N</td>\n",
       "      <td>1.7627</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-995227</td>\n",
       "      <td>H6PbCI3N</td>\n",
       "      <td>1.5940</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-977013</td>\n",
       "      <td>H6PbCI3N</td>\n",
       "      <td>2.4371</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id    pretty_formula  band_gap  nelements\n",
       "0   mp-995203          H6PbCI3N    1.7575          5\n",
       "1  mp-1194995  H36Pb3C10(I5N2)2    1.7045          5\n",
       "2   mp-995214          H6PbCI3N    1.7627          5\n",
       "3   mp-995227          H6PbCI3N    1.5940          5\n",
       "4   mp-977013          H6PbCI3N    2.4371          5"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mpr = MPRester(\"......\") \n",
    "materials = mpr.query(\"C-N-H-Pb-I\", [\"material_id\", \"pretty_formula\", \"band_gap\", \"nelements\"])\n",
    "\n",
    "\n",
    "HOIP = pd.DataFrame(materials)\n",
    "HOIP.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "df722de5",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for He. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Ne. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Ar. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Rf. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Db. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Sg. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Bh. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Hs. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Mt. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Ds. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Rg. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Cn. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Nh. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Fl. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Mc. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Lv. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Ts. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\miniconda3\\lib\\site-packages\\pymatgen\\core\\periodic_table.py:213: UserWarning: No electronegativity for Og. Setting to NaN. This has no physical meaning, and is mainly done to avoid errors caused by the code expecting a float.\n",
      "  warnings.warn(\n",
      "100%|█████████████████████████████████████████████████████████████████████████████| 2398/2398 [00:04<00:00, 502.22it/s]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>material_id</th>\n",
       "      <th>pretty_formula</th>\n",
       "      <th>band_gap</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-1001571</td>\n",
       "      <td>CaFeO3</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-1016821</td>\n",
       "      <td>BaSiO3</td>\n",
       "      <td>1.0186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-1016823</td>\n",
       "      <td>BaGeO3</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-1016828</td>\n",
       "      <td>CdIrO3</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-1016833</td>\n",
       "      <td>CaRhO3</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id pretty_formula  band_gap\n",
       "0  mp-1001571         CaFeO3    0.0000\n",
       "1  mp-1016821         BaSiO3    1.0186\n",
       "2  mp-1016823         BaGeO3    0.0000\n",
       "3  mp-1016828         CdIrO3    0.0000\n",
       "4  mp-1016833         CaRhO3    0.0000"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mpr = MPRester(\".....\") \n",
    "materials = mpr.query(\"**O3\", [\"material_id\", \"pretty_formula\", \"band_gap\"])\n",
    "materials\n",
    "\n",
    "perov = pd.DataFrame(materials)\n",
    "perov.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "992b2111",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 2398 entries, 0 to 2397\n",
      "Data columns (total 3 columns):\n",
      " #   Column          Non-Null Count  Dtype  \n",
      "---  ------          --------------  -----  \n",
      " 0   material_id     2398 non-null   object \n",
      " 1   pretty_formula  2398 non-null   object \n",
      " 2   band_gap        2398 non-null   float64\n",
      "dtypes: float64(1), object(2)\n",
      "memory usage: 56.3+ KB\n"
     ]
    }
   ],
   "source": [
    "perov.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "c27c6c43",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 261,
       "width": 381
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(perov[\"band_gap\"], bins=20)\n",
    "plt.xlabel(\"Band gap (eV)\")\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "ef43f12d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>material_id</th>\n",
       "      <th>band_gap</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-557591</td>\n",
       "      <td>5.6144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-557653</td>\n",
       "      <td>4.1306</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-639733</td>\n",
       "      <td>5.6305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-1249090</td>\n",
       "      <td>5.7815</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-554151</td>\n",
       "      <td>5.7147</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id  band_gap\n",
       "0   mp-557591    5.6144\n",
       "1   mp-557653    4.1306\n",
       "2   mp-639733    5.6305\n",
       "3  mp-1249090    5.7815\n",
       "4   mp-554151    5.7147"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "materials = mpr.query({\"chemsys\":\"O-Si\"}, ['material_id', \"band_gap\"])\n",
    "\n",
    "df = pd.DataFrame(materials)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "11cac334",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 372
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "with MPRester(\"......\") as mpr:\n",
    "    entries = mpr.query(\"Ta-C\", [\"material_id\", \"spacegroup.crystal_system\"])\n",
    "    systems = [entry[\"spacegroup.crystal_system\"] for entry in entries]\n",
    "    counts = {}\n",
    "    \n",
    "    for system in systems:\n",
    "        if system in counts:\n",
    "            counts[system] += 1\n",
    "        else:\n",
    "            counts[system] = 1\n",
    "            \n",
    "    plt.bar(range(len(counts)), counts.values())\n",
    "    plt.xticks(range(len(counts)), [key[0:7] for key in counts.keys()])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "cadaa1fb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "53\n",
      "KPO3\n",
      "5.0422\n",
      "{'structure': Structure Summary\n",
      "Lattice\n",
      "    abc : 4.622368 10.532778 14.283612732012022\n",
      " angles : 78.68805047946931 90.0 90.0\n",
      " volume : 681.9080032992589\n",
      "      A : 4.622368 0.0 0.0\n",
      "      B : 0.0 10.532778 0.0\n",
      "      C : 0.0 2.80174 14.006136\n",
      "PeriodicSite: K (1.2899, 11.7096, 9.3958) [0.2791, 0.9333, 0.6708]\n",
      "PeriodicSite: K (3.6011, 3.0258, 11.6134) [0.7791, 0.0667, 0.8292]\n",
      "PeriodicSite: K (3.3324, 1.6249, 4.6104) [0.7209, 0.0667, 0.3292]\n",
      "PeriodicSite: K (1.0213, 10.3087, 2.3927) [0.2209, 0.9333, 0.1708]\n",
      "PeriodicSite: K (3.4028, 8.1106, 8.8415) [0.7361, 0.6021, 0.6313]\n",
      "PeriodicSite: K (1.0916, 6.6248, 12.1677) [0.2361, 0.3979, 0.8687]\n",
      "PeriodicSite: K (1.2196, 5.2240, 5.1647) [0.2639, 0.3979, 0.3687]\n",
      "PeriodicSite: K (3.5308, 6.7097, 1.8384) [0.7639, 0.6021, 0.1313]\n",
      "PeriodicSite: P (3.6775, 3.8259, 7.7787) [0.7956, 0.2155, 0.5554]\n",
      "PeriodicSite: P (1.3663, 10.9095, 13.2305) [0.2956, 0.7845, 0.9446]\n",
      "PeriodicSite: P (0.9449, 9.5086, 6.2274) [0.2044, 0.7845, 0.4446]\n",
      "PeriodicSite: P (3.2561, 2.4251, 0.7757) [0.7044, 0.2155, 0.0554]\n",
      "PeriodicSite: P (1.3730, 5.0912, 9.1465) [0.2970, 0.3097, 0.6530]\n",
      "PeriodicSite: P (3.6842, 9.6442, 11.8627) [0.7970, 0.6903, 0.8470]\n",
      "PeriodicSite: P (3.2494, 8.2434, 4.8597) [0.7030, 0.6903, 0.3470]\n",
      "PeriodicSite: P (0.9382, 3.6903, 2.1434) [0.2030, 0.3097, 0.1530]\n",
      "PeriodicSite: O (0.5451, 4.4896, 7.8664) [0.1179, 0.2769, 0.5616]\n",
      "PeriodicSite: O (2.8563, 10.2458, 13.1428) [0.6179, 0.7231, 0.9384]\n",
      "PeriodicSite: O (4.0773, 8.8449, 6.1397) [0.8821, 0.7231, 0.4384]\n",
      "PeriodicSite: O (1.7661, 3.0887, 0.8634) [0.3821, 0.2769, 0.0616]\n",
      "PeriodicSite: O (2.8519, 5.0735, 8.4653) [0.6170, 0.3209, 0.6044]\n",
      "PeriodicSite: O (0.5407, 9.6619, 12.5439) [0.1170, 0.6791, 0.8956]\n",
      "PeriodicSite: O (1.7705, 8.2610, 5.5409) [0.3830, 0.6791, 0.3956]\n",
      "PeriodicSite: O (4.0817, 3.6726, 1.4622) [0.8830, 0.3209, 0.1044]\n",
      "PeriodicSite: O (1.2898, 4.1565, 10.3188) [0.2790, 0.1987, 0.7367]\n",
      "PeriodicSite: O (3.6009, 10.5789, 10.6904) [0.7790, 0.8013, 0.7633]\n",
      "PeriodicSite: O (3.3326, 9.1780, 3.6873) [0.7210, 0.8013, 0.2633]\n",
      "PeriodicSite: O (1.0214, 2.7556, 3.3158) [0.2210, 0.1987, 0.2367]\n",
      "PeriodicSite: O (1.0453, 6.5408, 9.3477) [0.2261, 0.4435, 0.6674]\n",
      "PeriodicSite: O (3.3564, 8.1946, 11.6615) [0.7261, 0.5565, 0.8326]\n",
      "PeriodicSite: O (3.5771, 6.7938, 4.6584) [0.7739, 0.5565, 0.3326]\n",
      "PeriodicSite: O (1.2659, 5.1399, 2.3447) [0.2739, 0.4435, 0.1674]\n",
      "PeriodicSite: O (3.3503, 1.2092, 1.6459) [0.7248, 0.0835, 0.1175]\n",
      "PeriodicSite: O (1.0392, 10.7245, 5.3572) [0.2248, 0.9165, 0.3825]\n",
      "PeriodicSite: O (1.2720, 12.1254, 12.3603) [0.2752, 0.9165, 0.8825]\n",
      "PeriodicSite: O (3.5832, 2.6100, 8.6489) [0.7752, 0.0835, 0.6175]\n",
      "PeriodicSite: O (1.0516, 8.1759, 0.6872) [0.2275, 0.7632, 0.0491]\n",
      "PeriodicSite: O (3.3627, 3.7577, 6.3158) [0.7275, 0.2368, 0.4509]\n",
      "PeriodicSite: O (3.5708, 5.1586, 13.3189) [0.7725, 0.2368, 0.9509]\n",
      "PeriodicSite: O (1.2596, 9.5768, 7.6903) [0.2725, 0.7632, 0.5491], 'material_id': 'mp-5039', 'pretty_formula': 'KPO3', 'e_above_hull': 0, 'band_gap': 5.0422, 'band_structure': None}\n"
     ]
    }
   ],
   "source": [
    "mpr = MPRester(\".....\")\n",
    "\n",
    "criteria = {\"elements\":{\"$in\":[\"Li\", \"Na\", \"K\"], \"$all\":[\"O\"]}, # All compounds contain O, and must have either Li, Na, or K \n",
    "           \"nelements\":3,\n",
    "           \"icsd_ids\": {\"$gte\":0},  # greater than or equal to \"0\"\n",
    "           \"e_above_hull\": {\"$lte\": 0.01},   # less than or equal to \"0.01\"\n",
    "            \"anonymous_formula\": {\"A\":1, \"B\":1, \"C\":3},  # ABO3\n",
    "            \"band_gap\": {\"$gt\":1}  # greater than 1\n",
    "           }\n",
    "\n",
    "props = [\"structure\", \"material_id\", \"pretty_formula\", \"e_above_hull\", \"band_gap\",\n",
    "        \"band_structure\"]\n",
    "entries = mpr.query(criteria=criteria, properties=props)\n",
    "\n",
    "print(len(entries))\n",
    "\n",
    "for e in entries:\n",
    "    print(e[\"pretty_formula\"])\n",
    "    print(e[\"band_gap\"])\n",
    "    print(e)\n",
    "    break"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c6d51493",
   "metadata": {},
   "source": [
    "# Viewing Si3N4 structure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "ec3dbd9b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d6cc2a75f3574480b732e1b16d831869",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9f8df6e0979d4442a14113c75dd224bf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "NGLWidget()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from pymatgen.io.cif import CifParser\n",
    "import nglview as nv\n",
    "from nglview import show_structure_file\n",
    "\n",
    "name = 'mp-697331'\n",
    "mpr = MPRester(\".....\")\n",
    "structure = mpr.get_structures(name)\n",
    "# print(structure)\n",
    "\n",
    "structure_cif = mpr.get_data(name, data_type='', prop='cif').get('cif')\n",
    "# print(structure_cif)\n",
    "\n",
    "# SAve the cif file to disc\n",
    "with open(\"{}.cif\".format(name), \"w\") as output:\n",
    "    output.write(structure_cif)\n",
    "    \n",
    "# Visualise\n",
    "view = nv.show_structure_file(\"{}.cif\".format(name))\n",
    "view._remote_call('setSize', args = ['', '600px'])\n",
    "view.camera = 'orthographic'\n",
    "view.add_unitcell()\n",
    "\n",
    "# save\n",
    "# view.download_image(\"{}.png\".format(name))\n",
    "\n",
    "view"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "1e5f9d55",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Structure Summary\n",
       " Lattice\n",
       "     abc : 7.882683366390471 7.882683366390471 6.500556316212483\n",
       "  angles : 68.68079415898275 68.68079415898275 48.0063465923498\n",
       "  volume : 275.4047791632832\n",
       "       A : 3.206575 7.201012 0.0\n",
       "       B : -3.206575 7.201012 0.0\n",
       "       C : 0.0 2.587089 5.963573\n",
       " PeriodicSite: Mg (-0.1024, 3.6030, 0.5300) [0.2182, 0.2502, 0.0889]\n",
       " PeriodicSite: Mg (-0.1024, 12.0926, 2.4517) [0.7498, 0.7818, 0.4111]\n",
       " PeriodicSite: Mg (0.1024, 13.3862, 5.4335) [0.7818, 0.7498, 0.9111]\n",
       " PeriodicSite: Mg (0.1024, 4.8965, 3.5118) [0.2502, 0.2182, 0.5889]\n",
       " PeriodicSite: Mg (-3.0228, 7.8478, 1.4909) [0.0287, 0.9713, 0.2500]\n",
       " PeriodicSite: Mg (3.0228, 9.1413, 4.4727) [0.9713, 0.0287, 0.7500]\n",
       " PeriodicSite: Mg (0.2007, 7.8478, 1.4909) [0.5313, 0.4687, 0.2500]\n",
       " PeriodicSite: Mg (-0.2007, 9.1413, 4.4727) [0.4687, 0.5313, 0.7500]\n",
       " PeriodicSite: Ni (-1.7455, 5.6643, 0.4586) [0.1073, 0.6517, 0.0769]\n",
       " PeriodicSite: Ni (-1.7455, 10.0312, 2.5232) [0.3483, 0.8927, 0.4231]\n",
       " PeriodicSite: Ni (1.7455, 11.3248, 5.5050) [0.8927, 0.3483, 0.9231]\n",
       " PeriodicSite: Ni (1.7455, 6.9579, 3.4403) [0.6517, 0.1073, 0.5769]\n",
       " PeriodicSite: H (-1.2738, 4.9663, 1.7905) [0.0923, 0.4895, 0.3002]\n",
       " PeriodicSite: H (-1.2738, 10.7293, 1.1912) [0.5105, 0.9077, 0.1998]\n",
       " PeriodicSite: H (1.2738, 12.0228, 4.1730) [0.9077, 0.5105, 0.6998]\n",
       " PeriodicSite: H (1.2738, 6.2599, 4.7723) [0.4895, 0.0923, 0.8002]\n",
       " PeriodicSite: H (-1.1346, 7.4648, 5.2307) [0.1838, 0.5377, 0.8771]\n",
       " PeriodicSite: H (-1.1346, 10.8178, 3.7146) [0.4623, 0.8162, 0.6229]\n",
       " PeriodicSite: H (1.1346, 9.5243, 0.7329) [0.8162, 0.4623, 0.1229]\n",
       " PeriodicSite: H (1.1346, 6.1713, 2.2489) [0.5377, 0.1838, 0.3771]\n",
       " PeriodicSite: H (-1.3320, 7.1659, 0.3519) [0.2793, 0.6947, 0.0590]\n",
       " PeriodicSite: H (-1.3320, 8.5297, 2.6299) [0.3053, 0.7207, 0.4410]\n",
       " PeriodicSite: H (1.3320, 9.8233, 5.6117) [0.7207, 0.3053, 0.9410]\n",
       " PeriodicSite: H (1.3320, 8.4594, 3.3337) [0.6947, 0.2793, 0.5590]\n",
       " PeriodicSite: H (-0.0807, 12.7830, 0.3980) [0.8630, 0.8882, 0.0667]\n",
       " PeriodicSite: H (-0.0807, 2.9126, 2.5838) [0.1118, 0.1370, 0.4333]\n",
       " PeriodicSite: H (0.0807, 4.2061, 5.5656) [0.1370, 0.1118, 0.9333]\n",
       " PeriodicSite: H (0.0807, 14.0765, 3.3798) [0.8882, 0.8630, 0.5667]]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "structure"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "98190d9b",
   "metadata": {},
   "source": [
    "# Matminer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "bb7aac5d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "boltztrap_mp: Effective mass and thermoelectric properties of 8924 compounds in The  Materials Project database that are calculated by the BoltzTraP software package run on the GGA-PBE or GGA+U density functional theory calculation results. The properties are reported at the temperature of 300 Kelvin and the carrier concentration of 1e18 1/cm3.\n",
      "\n",
      "brgoch_superhard_training: 2574 materials used for training regressors that predict shear and bulk modulus.\n",
      "\n",
      "castelli_perovskites: 18,928 perovskites generated with ABX combinatorics, calculating gllbsc band gap and pbe structure, and also reporting absolute band edge positions and heat of formation.\n",
      "\n",
      "citrine_thermal_conductivity: Thermal conductivity of 872 compounds measured experimentally and retrieved from Citrine database from various references. The reported values are measured at various temperatures of which 295 are at room temperature.\n",
      "\n",
      "dielectric_constant: 1,056 structures with dielectric properties, calculated with DFPT-PBE.\n",
      "\n",
      "double_perovskites_gap: Band gap of 1306 double perovskites (a_1-b_1-a_2-b_2-O6) calculated using ﻿Gritsenko, van Leeuwen, van Lenthe and Baerends potential (gllbsc) in GPAW.\n",
      "\n",
      "double_perovskites_gap_lumo: Supplementary lumo data of 55 atoms for the double_perovskites_gap dataset.\n",
      "\n",
      "elastic_tensor_2015: 1,181 structures with elastic properties calculated with DFT-PBE.\n",
      "\n",
      "expt_formation_enthalpy: Experimental formation enthalpies for inorganic compounds, collected from years of calorimetric experiments. There are 1,276 entries in this dataset, mostly binary compounds. Matching mpids or oqmdids as well as the DFT-computed formation energies are also added (if any).\n",
      "\n",
      "expt_formation_enthalpy_kingsbury: Dataset containing experimental standard formation enthalpies for solids. Formation enthalpies were compiled primarily from Kim et al., Kubaschewski, and the NIST JANAF tables (see references). Elements, liquids, and gases were excluded. Data were deduplicated such that each material is associated with a single formation enthalpy value. Refer to Wang et al. (see references) for a complete desciption of the methods used. Materials Project database IDs (mp-ids) were assigned to materials from among computed materials in the Materials Project database (version 2021.03.22) that were 1) not marked 'theoretical', 2) had structures matching at least one ICSD material, and 3) were within 200 meV of the DFT-computed stable energy hull (e_above_hull < 0.2 eV). Among these candidates, we chose the mp-id with the lowest e_above_hull that matched the reported spacegroup (where available).\n",
      "\n",
      "expt_gap: Experimental band gap of 6354 inorganic semiconductors.\n",
      "\n",
      "expt_gap_kingsbury: Identical to the matbench_expt_gap dataset, except that Materials Project database IDs (mp-ids) have been associated with each material using the same method as described for the expt_formation_enthalpy_kingsbury dataset. Columns have also been renamed for consistency with the formation enthalpy data.\n",
      "\n",
      "flla: 3938 structures and computed formation energies from \"Crystal Structure Representations for Machine Learning Models of Formation Energies.\"\n",
      "\n",
      "glass_binary: Metallic glass formation data for binary alloys, collected from various experimental techniques such as melt-spinning or mechanical alloying. This dataset covers all compositions with an interval of 5 at. % in 59 binary systems, containing a total of 5959 alloys in the dataset. The target property of this dataset is the glass forming ability (GFA), i.e. whether the composition can form monolithic glass or not, which is either 1 for glass forming or 0 for non-full glass forming.\n",
      "\n",
      "glass_binary_v2: Identical to glass_binary dataset, but with duplicate entries merged. If there was a disagreement in gfa when merging the class was defaulted to 1.\n",
      "\n",
      "glass_ternary_hipt: Metallic glass formation dataset for ternary alloys, collected from the high-throughput sputtering experiments measuring whether it is possible to form a glass using sputtering. The hipt experimental data are of the Co-Fe-Zr, Co-Ti-Zr, Co-V-Zr and Fe-Ti-Nb ternary systems.\n",
      "\n",
      "glass_ternary_landolt: Metallic glass formation dataset for ternary alloys, collected from the \"Nonequilibrium Phase Diagrams of Ternary Amorphous Alloys,’ a volume of the Landolt– Börnstein collection. This dataset contains experimental measurements of whether it is possible to form a glass using a variety of processing techniques at thousands of compositions from hundreds of ternary systems. The processing techniques are designated in the \"processing\" column. There are originally 7191 experiments in this dataset, will be reduced to 6203 after deduplicated, and will be further reduced to 6118 if combining multiple data for one composition. There are originally 6780 melt-spinning experiments in this dataset, will be reduced to 5800 if deduplicated, and will be further reduced to 5736 if combining multiple experimental data for one composition.\n",
      "\n",
      "heusler_magnetic: 1153 Heusler alloys with DFT-calculated magnetic and electronic properties. The 1153 alloys include 576 full, 449 half and 128 inverse Heusler alloys. The data are extracted and cleaned (including de-duplicating) from Citrine.\n",
      "\n",
      "jarvis_dft_2d: Various properties of 636 2D materials computed with the OptB88vdW and TBmBJ functionals taken from the JARVIS DFT database.\n",
      "\n",
      "jarvis_dft_3d: Various properties of 25,923 bulk materials computed with the OptB88vdW and TBmBJ functionals taken from the JARVIS DFT database.\n",
      "\n",
      "jarvis_ml_dft_training: Various properties of 24,759 bulk and 2D materials computed with the OptB88vdW and TBmBJ functionals taken from the JARVIS DFT database.\n",
      "\n",
      "m2ax: Elastic properties of 223 stable M2AX compounds from \"A comprehensive survey of M2AX phase elastic properties\" by Cover et al. Calculations are PAW PW91.\n",
      "\n",
      "matbench_dielectric: Matbench v0.1 test dataset for predicting refractive index from structure. Adapted from Materials Project database. Removed entries having a formation energy (or energy above the convex hull) more than 150meV and those having refractive indices less than 1 and those containing noble gases. Retrieved April 2, 2019. For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_expt_gap: Matbench v0.1 test dataset for predicting experimental band gap from composition alone. Retrieved from Zhuo et al. supplementary information. Deduplicated according to composition, removing compositions with reported band gaps spanning more than a 0.1eV range; remaining compositions were assigned values based on the closest experimental value to the mean experimental value for that composition among all reports. For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_expt_is_metal: Matbench v0.1 test dataset for classifying metallicity from composition alone. Retrieved from Zhuo et al. supplementary information. Deduplicated according to composition, ensuring no conflicting reports were entered for any compositions (i.e., no reported compositions were both metal and nonmetal). For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_glass: Matbench v0.1 test dataset for predicting full bulk metallic glass formation ability from chemical formula. Retrieved from \"Nonequilibrium Phase Diagrams of Ternary Amorphous Alloys,’ a volume of the Landolt– Börnstein collection. Deduplicated according to composition, ensuring no compositions were reported as both GFA and not GFA (i.e., all reports agreed on the classification designation). For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_jdft2d: Matbench v0.1 test dataset for predicting exfoliation energies from crystal structure (computed with the OptB88vdW and TBmBJ functionals). Adapted from the JARVIS DFT database. For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_log_gvrh: Matbench v0.1 test dataset for predicting DFT log10 VRH-average shear modulus from structure. Adapted from Materials Project database. Removed entries having a formation energy (or energy above the convex hull) more than 150meV and those having negative G_Voigt, G_Reuss, G_VRH, K_Voigt, K_Reuss, or K_VRH and those failing G_Reuss <= G_VRH <= G_Voigt or K_Reuss <= K_VRH <= K_Voigt and those containing noble gases. Retrieved April 2, 2019. For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_log_kvrh: Matbench v0.1 test dataset for predicting DFT log10 VRH-average bulk modulus from structure. Adapted from Materials Project database. Removed entries having a formation energy (or energy above the convex hull) more than 150meV and those having negative G_Voigt, G_Reuss, G_VRH, K_Voigt, K_Reuss, or K_VRH and those failing G_Reuss <= G_VRH <= G_Voigt or K_Reuss <= K_VRH <= K_Voigt and those containing noble gases. Retrieved April 2, 2019. For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_mp_e_form: Matbench v0.1 test dataset for predicting DFT formation energy from structure. Adapted from Materials Project database. Removed entries having formation energy more than 2.5eV and those containing noble gases. Retrieved April 2, 2019. For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_mp_gap: Matbench v0.1 test dataset for predicting DFT PBE band gap from structure. Adapted from Materials Project database. Removed entries having a formation energy (or energy above the convex hull) more than 150meV and those containing noble gases. Retrieved April 2, 2019. For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_mp_is_metal: Matbench v0.1 test dataset for predicting DFT metallicity from structure. Adapted from Materials Project database. Removed entries having a formation energy (or energy above the convex hull) more than 150meV and those containing noble gases. Retrieved April 2, 2019. For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_perovskites: Matbench v0.1 test dataset for predicting formation energy from crystal structure. Adapted from an original dataset generated by Castelli et al. For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_phonons: Matbench v0.1 test dataset for predicting vibration properties from crystal structure. Original data retrieved from Petretto et al. Original calculations done via ABINIT in the harmonic approximation based on density functional perturbation theory. Removed entries having a formation energy (or energy above the convex hull) more than 150meV. For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "matbench_steels: Matbench v0.1 test dataset for predicting steel yield strengths from chemical composition alone. Retrieved from Citrine informatics. Deduplicated. For benchmarking w/ nested cross validation, the order of the dataset must be identical to the retrieved data; refer to the Automatminer/Matbench publication for more details.\n",
      "\n",
      "mp_all_20181018: A complete copy of the Materials Project database as of 10/18/2018. mp_all files contain structure data for each material while mp_nostruct does not.\n",
      "\n",
      "mp_nostruct_20181018: A complete copy of the Materials Project database as of 10/18/2018. mp_all files contain structure data for each material while mp_nostruct does not.\n",
      "\n",
      "phonon_dielectric_mp: Phonon (lattice/atoms vibrations) and dielectric properties of 1296 compounds computed via ABINIT software package in the harmonic approximation based on density functional perturbation theory.\n",
      "\n",
      "piezoelectric_tensor: 941 structures with piezoelectric properties, calculated with DFT-PBE.\n",
      "\n",
      "ricci_boltztrap_mp_tabular: Ab-initio electronic transport database for inorganic materials. Complex multivariable BoltzTraP simulation data is condensed down into tabular form of two main motifs: average eigenvalues at set moderate carrier concentrations and temperatures, and optimal values among all carrier concentrations and temperatures within certain ranges. Here are reported the average of the eigenvalues of conductivity effective mass (mₑᶜᵒⁿᵈ), the Seebeck coefficient (S), the conductivity (σ), the electronic thermal conductivity (κₑ), and the Power Factor (PF) at a doping level of 10¹⁸ cm⁻³ and at a temperature of 300 K for n- and p-type. Also, the maximum values for S, σ, PF, and the minimum value for κₑ chosen among the temperatures [100, 1300] K, the doping levels [10¹⁶, 10²¹] cm⁻³, and doping types are reported. The properties that depend on the relaxation time are reported divided by the constant value 10⁻¹⁴. The average of the eigenvalues for all the properties at all the temperatures, doping levels, and doping types are reported in the tables for each entry. Data is indexed by materials project id (mpid)\n",
      "\n",
      "steel_strength: 312 steels with experimental yield strength and ultimate tensile strength, extracted and cleaned (including de-duplicating) from Citrine.\n",
      "\n",
      "superconductivity2018: Dataset of ~16,000 experimental superconductivity records (critical temperatures) from Stanev et al., originally from the Japanese National Institute for Materials Science. Does not include structural data. Includes ~300 measurements from materials found without superconductivity (Tc=0). No modifications were made to the core dataset, aside from basic file type change to json for (un)packaging with matminer. Reproduced under the Creative Commons 4.0 license, which can be found here: http://creativecommons.org/licenses/by/4.0/.\n",
      "\n",
      "wolverton_oxides: 4,914 perovskite oxides containing composition data, lattice constants, and formation + vacancy formation energies. All perovskites are of the form ABO3. Adapted from a dataset presented by Emery and Wolverton.\n",
      "\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['boltztrap_mp',\n",
       " 'brgoch_superhard_training',\n",
       " 'castelli_perovskites',\n",
       " 'citrine_thermal_conductivity',\n",
       " 'dielectric_constant',\n",
       " 'double_perovskites_gap',\n",
       " 'double_perovskites_gap_lumo',\n",
       " 'elastic_tensor_2015',\n",
       " 'expt_formation_enthalpy',\n",
       " 'expt_formation_enthalpy_kingsbury',\n",
       " 'expt_gap',\n",
       " 'expt_gap_kingsbury',\n",
       " 'flla',\n",
       " 'glass_binary',\n",
       " 'glass_binary_v2',\n",
       " 'glass_ternary_hipt',\n",
       " 'glass_ternary_landolt',\n",
       " 'heusler_magnetic',\n",
       " 'jarvis_dft_2d',\n",
       " 'jarvis_dft_3d',\n",
       " 'jarvis_ml_dft_training',\n",
       " 'm2ax',\n",
       " 'matbench_dielectric',\n",
       " 'matbench_expt_gap',\n",
       " 'matbench_expt_is_metal',\n",
       " 'matbench_glass',\n",
       " 'matbench_jdft2d',\n",
       " 'matbench_log_gvrh',\n",
       " 'matbench_log_kvrh',\n",
       " 'matbench_mp_e_form',\n",
       " 'matbench_mp_gap',\n",
       " 'matbench_mp_is_metal',\n",
       " 'matbench_perovskites',\n",
       " 'matbench_phonons',\n",
       " 'matbench_steels',\n",
       " 'mp_all_20181018',\n",
       " 'mp_nostruct_20181018',\n",
       " 'phonon_dielectric_mp',\n",
       " 'piezoelectric_tensor',\n",
       " 'ricci_boltztrap_mp_tabular',\n",
       " 'steel_strength',\n",
       " 'superconductivity2018',\n",
       " 'wolverton_oxides']"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from matminer.datasets import get_available_datasets\n",
    "from pymatgen.core import Composition, Element, bonds\n",
    "import pandas as pd\n",
    "\n",
    "get_available_datasets()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "7c2ac10e",
   "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>epsilon_x opt</th>\n",
       "      <th>epsilon_y opt</th>\n",
       "      <th>epsilon_z opt</th>\n",
       "      <th>exfoliation_en</th>\n",
       "      <th>structure</th>\n",
       "      <th>e_form</th>\n",
       "      <th>structure initial</th>\n",
       "      <th>jid</th>\n",
       "      <th>gap tbmbj</th>\n",
       "      <th>epsilon_x tbmbj</th>\n",
       "      <th>epsilon_y tbmbj</th>\n",
       "      <th>epsilon_z tbmbj</th>\n",
       "      <th>mpid</th>\n",
       "      <th>gap opt</th>\n",
       "      <th>composition</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10.3295</td>\n",
       "      <td>10.3295</td>\n",
       "      <td>7.0367</td>\n",
       "      <td>63.593833</td>\n",
       "      <td>[[1.49323139 3.32688406 7.26257785] Hf, [3.326...</td>\n",
       "      <td>-0.624</td>\n",
       "      <td>[[1.49180494 3.32370494 7.25734791] Hf, [3.323...</td>\n",
       "      <td>JVASP-19990</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>mp-13963</td>\n",
       "      <td>0.0341</td>\n",
       "      <td>(Hf, Si, Te)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.7325</td>\n",
       "      <td>5.6477</td>\n",
       "      <td>2.6267</td>\n",
       "      <td>134.863750</td>\n",
       "      <td>[[1.85068084 4.37698238 6.9301577 ] As, [0.   ...</td>\n",
       "      <td>0.160</td>\n",
       "      <td>[[1.85325003 4.36491291 6.93222312] As, [0.   ...</td>\n",
       "      <td>JVASP-31349</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>mp-158</td>\n",
       "      <td>0.7467</td>\n",
       "      <td>(As)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6.4357</td>\n",
       "      <td>5.2377</td>\n",
       "      <td>3.0070</td>\n",
       "      <td>43.114667</td>\n",
       "      <td>[[ 0.          2.0213325  11.97279555] Ti, [ 1...</td>\n",
       "      <td>-2.231</td>\n",
       "      <td>[[ 0.          2.0239455  11.96404455] Ti, [ 1...</td>\n",
       "      <td>JVASP-27890</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>mp-23002</td>\n",
       "      <td>0.0023</td>\n",
       "      <td>(Ti, Br, O)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9.3915</td>\n",
       "      <td>9.3915</td>\n",
       "      <td>4.2669</td>\n",
       "      <td>240.715488</td>\n",
       "      <td>[[2.39882726 2.39882726 2.53701553] In, [0.054...</td>\n",
       "      <td>0.240</td>\n",
       "      <td>[[2.53701507 2.53701507 2.53701604] In, [0.057...</td>\n",
       "      <td>JVASP-31353</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>mp-23313</td>\n",
       "      <td>0.0266</td>\n",
       "      <td>(In, Bi)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.7367</td>\n",
       "      <td>2.7367</td>\n",
       "      <td>2.0162</td>\n",
       "      <td>67.442833</td>\n",
       "      <td>[[-1.83484554e-06  1.73300105e+00  2.61675943e...</td>\n",
       "      <td>-2.245</td>\n",
       "      <td>[[-2.000000e-06  1.731207e+00  2.616762e+00] N...</td>\n",
       "      <td>JVASP-31356</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>mp-25214</td>\n",
       "      <td>0.0484</td>\n",
       "      <td>(Nb, O)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   epsilon_x opt  epsilon_y opt  epsilon_z opt  exfoliation_en  \\\n",
       "0        10.3295        10.3295         7.0367       63.593833   \n",
       "1         4.7325         5.6477         2.6267      134.863750   \n",
       "2         6.4357         5.2377         3.0070       43.114667   \n",
       "3         9.3915         9.3915         4.2669      240.715488   \n",
       "4         2.7367         2.7367         2.0162       67.442833   \n",
       "\n",
       "                                           structure  e_form  \\\n",
       "0  [[1.49323139 3.32688406 7.26257785] Hf, [3.326...  -0.624   \n",
       "1  [[1.85068084 4.37698238 6.9301577 ] As, [0.   ...   0.160   \n",
       "2  [[ 0.          2.0213325  11.97279555] Ti, [ 1...  -2.231   \n",
       "3  [[2.39882726 2.39882726 2.53701553] In, [0.054...   0.240   \n",
       "4  [[-1.83484554e-06  1.73300105e+00  2.61675943e...  -2.245   \n",
       "\n",
       "                                   structure initial          jid  gap tbmbj  \\\n",
       "0  [[1.49180494 3.32370494 7.25734791] Hf, [3.323...  JVASP-19990        NaN   \n",
       "1  [[1.85325003 4.36491291 6.93222312] As, [0.   ...  JVASP-31349        NaN   \n",
       "2  [[ 0.          2.0239455  11.96404455] Ti, [ 1...  JVASP-27890        NaN   \n",
       "3  [[2.53701507 2.53701507 2.53701604] In, [0.057...  JVASP-31353        NaN   \n",
       "4  [[-2.000000e-06  1.731207e+00  2.616762e+00] N...  JVASP-31356        NaN   \n",
       "\n",
       "   epsilon_x tbmbj  epsilon_y tbmbj  epsilon_z tbmbj      mpid  gap opt  \\\n",
       "0              NaN              NaN              NaN  mp-13963   0.0341   \n",
       "1              NaN              NaN              NaN    mp-158   0.7467   \n",
       "2              NaN              NaN              NaN  mp-23002   0.0023   \n",
       "3              NaN              NaN              NaN  mp-23313   0.0266   \n",
       "4              NaN              NaN              NaN  mp-25214   0.0484   \n",
       "\n",
       "    composition  \n",
       "0  (Hf, Si, Te)  \n",
       "1          (As)  \n",
       "2   (Ti, Br, O)  \n",
       "3      (In, Bi)  \n",
       "4       (Nb, O)  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from matminer.datasets import load_dataset\n",
    "\n",
    "df = load_dataset(\"jarvis_dft_2d\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "1a28341f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>material_id</th>\n",
       "      <th>formula</th>\n",
       "      <th>nsites</th>\n",
       "      <th>space_group</th>\n",
       "      <th>volume</th>\n",
       "      <th>structure</th>\n",
       "      <th>band_gap</th>\n",
       "      <th>e_electronic</th>\n",
       "      <th>e_total</th>\n",
       "      <th>n</th>\n",
       "      <th>poly_electronic</th>\n",
       "      <th>poly_total</th>\n",
       "      <th>pot_ferroelectric</th>\n",
       "      <th>cif</th>\n",
       "      <th>meta</th>\n",
       "      <th>poscar</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-441</td>\n",
       "      <td>Rb2Te</td>\n",
       "      <td>3</td>\n",
       "      <td>225</td>\n",
       "      <td>159.501208</td>\n",
       "      <td>[[1.75725875 1.2425695  3.04366125] Rb, [5.271...</td>\n",
       "      <td>1.88</td>\n",
       "      <td>[[3.44115795, -3.097e-05, -6.276e-05], [-2.837...</td>\n",
       "      <td>[[6.23414745, -0.00035252, -9.796e-05], [-0.00...</td>\n",
       "      <td>1.86</td>\n",
       "      <td>3.44</td>\n",
       "      <td>6.23</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Rb2 Te1\\n1.0\\n5.271776 0.000000 3.043661\\n1.75...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-22881</td>\n",
       "      <td>CdCl2</td>\n",
       "      <td>3</td>\n",
       "      <td>166</td>\n",
       "      <td>84.298097</td>\n",
       "      <td>[[0. 0. 0.] Cd, [ 4.27210959  2.64061969 13.13...</td>\n",
       "      <td>3.52</td>\n",
       "      <td>[[3.34688382, -0.04498543, -0.22379197], [-0.0...</td>\n",
       "      <td>[[7.97018673, -0.29423886, -1.463590159999999]...</td>\n",
       "      <td>1.78</td>\n",
       "      <td>3.16</td>\n",
       "      <td>6.73</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Cd1 Cl2\\n1.0\\n3.850977 0.072671 5.494462\\n1.78...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-28013</td>\n",
       "      <td>MnI2</td>\n",
       "      <td>3</td>\n",
       "      <td>164</td>\n",
       "      <td>108.335875</td>\n",
       "      <td>[[0. 0. 0.] Mn, [-2.07904300e-06  2.40067320e+...</td>\n",
       "      <td>1.17</td>\n",
       "      <td>[[5.5430849, -5.28e-06, -2.5030000000000003e-0...</td>\n",
       "      <td>[[13.80606079, 0.0006911900000000001, 9.655e-0...</td>\n",
       "      <td>2.23</td>\n",
       "      <td>4.97</td>\n",
       "      <td>10.64</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Mn1 I2\\n1.0\\n4.158086 0.000000 0.000000\\n-2.07...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-567290</td>\n",
       "      <td>LaN</td>\n",
       "      <td>4</td>\n",
       "      <td>186</td>\n",
       "      <td>88.162562</td>\n",
       "      <td>[[-1.73309900e-06  2.38611186e+00  5.95256328e...</td>\n",
       "      <td>1.12</td>\n",
       "      <td>[[7.09316738, 7.99e-06, -0.0003864700000000000...</td>\n",
       "      <td>[[16.79535386, 8.199999999999997e-07, -0.00948...</td>\n",
       "      <td>2.65</td>\n",
       "      <td>7.04</td>\n",
       "      <td>17.99</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>La2 N2\\n1.0\\n4.132865 0.000000 0.000000\\n-2.06...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-560902</td>\n",
       "      <td>MnF2</td>\n",
       "      <td>6</td>\n",
       "      <td>136</td>\n",
       "      <td>82.826401</td>\n",
       "      <td>[[1.677294 2.484476 2.484476] Mn, [0. 0. 0.] M...</td>\n",
       "      <td>2.87</td>\n",
       "      <td>[[2.4239622, 7.452000000000001e-05, 6.06100000...</td>\n",
       "      <td>[[6.44055613, 0.0020446600000000002, 0.0013203...</td>\n",
       "      <td>1.53</td>\n",
       "      <td>2.35</td>\n",
       "      <td>7.12</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLDAUTYPE ...</td>\n",
       "      <td>Mn2 F4\\n1.0\\n3.354588 0.000000 0.000000\\n0.000...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id formula  nsites  space_group      volume  \\\n",
       "0      mp-441   Rb2Te       3          225  159.501208   \n",
       "1    mp-22881   CdCl2       3          166   84.298097   \n",
       "2    mp-28013    MnI2       3          164  108.335875   \n",
       "3   mp-567290     LaN       4          186   88.162562   \n",
       "4   mp-560902    MnF2       6          136   82.826401   \n",
       "\n",
       "                                           structure  band_gap  \\\n",
       "0  [[1.75725875 1.2425695  3.04366125] Rb, [5.271...      1.88   \n",
       "1  [[0. 0. 0.] Cd, [ 4.27210959  2.64061969 13.13...      3.52   \n",
       "2  [[0. 0. 0.] Mn, [-2.07904300e-06  2.40067320e+...      1.17   \n",
       "3  [[-1.73309900e-06  2.38611186e+00  5.95256328e...      1.12   \n",
       "4  [[1.677294 2.484476 2.484476] Mn, [0. 0. 0.] M...      2.87   \n",
       "\n",
       "                                        e_electronic  \\\n",
       "0  [[3.44115795, -3.097e-05, -6.276e-05], [-2.837...   \n",
       "1  [[3.34688382, -0.04498543, -0.22379197], [-0.0...   \n",
       "2  [[5.5430849, -5.28e-06, -2.5030000000000003e-0...   \n",
       "3  [[7.09316738, 7.99e-06, -0.0003864700000000000...   \n",
       "4  [[2.4239622, 7.452000000000001e-05, 6.06100000...   \n",
       "\n",
       "                                             e_total     n  poly_electronic  \\\n",
       "0  [[6.23414745, -0.00035252, -9.796e-05], [-0.00...  1.86             3.44   \n",
       "1  [[7.97018673, -0.29423886, -1.463590159999999]...  1.78             3.16   \n",
       "2  [[13.80606079, 0.0006911900000000001, 9.655e-0...  2.23             4.97   \n",
       "3  [[16.79535386, 8.199999999999997e-07, -0.00948...  2.65             7.04   \n",
       "4  [[6.44055613, 0.0020446600000000002, 0.0013203...  1.53             2.35   \n",
       "\n",
       "   poly_total  pot_ferroelectric  \\\n",
       "0        6.23              False   \n",
       "1        6.73              False   \n",
       "2       10.64              False   \n",
       "3       17.99              False   \n",
       "4        7.12              False   \n",
       "\n",
       "                                                 cif  \\\n",
       "0  #\\#CIF1.1\\n###################################...   \n",
       "1  #\\#CIF1.1\\n###################################...   \n",
       "2  #\\#CIF1.1\\n###################################...   \n",
       "3  #\\#CIF1.1\\n###################################...   \n",
       "4  #\\#CIF1.1\\n###################################...   \n",
       "\n",
       "                                                meta  \\\n",
       "0  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "1  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "2  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "3  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "4  {u'incar': u'NELM = 100\\nIBRION = 8\\nLDAUTYPE ...   \n",
       "\n",
       "                                              poscar  \n",
       "0  Rb2 Te1\\n1.0\\n5.271776 0.000000 3.043661\\n1.75...  \n",
       "1  Cd1 Cl2\\n1.0\\n3.850977 0.072671 5.494462\\n1.78...  \n",
       "2  Mn1 I2\\n1.0\\n4.158086 0.000000 0.000000\\n-2.07...  \n",
       "3  La2 N2\\n1.0\\n4.132865 0.000000 0.000000\\n-2.06...  \n",
       "4  Mn2 F4\\n1.0\\n3.354588 0.000000 0.000000\\n0.000...  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from matminer.datasets import load_dataset\n",
    "\n",
    "df = load_dataset(\"dielectric_constant\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "899b5ae1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset: dielectric_constant\n",
      "Description: 1,056 structures with dielectric properties, calculated with DFPT-PBE.\n",
      "Columns:\n",
      "\tband_gap: Measure of the conductivity of a material\n",
      "\tcif: optional: Description string for structure\n",
      "\te_electronic: electronic contribution to dielectric tensor\n",
      "\te_total: Total dielectric tensor incorporating both electronic and ionic contributions\n",
      "\tformula: Chemical formula of the material\n",
      "\tmaterial_id: Materials Project ID of the material\n",
      "\tmeta: optional, metadata descriptor of the datapoint\n",
      "\tn: Refractive Index\n",
      "\tnsites: The \\# of atoms in the unit cell of the calculation.\n",
      "\tpoly_electronic: the average of the eigenvalues of the electronic contribution to the dielectric tensor\n",
      "\tpoly_total: the average of the eigenvalues of the total (electronic and ionic) contributions to the dielectric tensor\n",
      "\tposcar: optional: Poscar metadata\n",
      "\tpot_ferroelectric: Whether the material is potentially ferroelectric\n",
      "\tspace_group: Integer specifying the crystallographic structure of the material\n",
      "\tstructure: pandas Series defining the structure of the material\n",
      "\tvolume: Volume of the unit cell in cubic angstroms, For supercell calculations, this quantity refers to the volume of the full supercell. \n",
      "Num Entries: 1056\n",
      "Reference: Petousis, I., Mrdjenovich, D., Ballouz, E., Liu, M., Winston, D.,\n",
      "Chen, W., Graf, T., Schladt, T. D., Persson, K. A. & Prinz, F. B.\n",
      "High-throughput screening of inorganic compounds for the discovery\n",
      "of novel dielectric and optical materials. Sci. Data 4, 160134 (2017).\n",
      "Bibtex citations: ['@Article{Petousis2017,\\nauthor={Petousis, Ioannis and Mrdjenovich, David and Ballouz, Eric\\nand Liu, Miao and Winston, Donald and Chen, Wei and Graf, Tanja\\nand Schladt, Thomas D. and Persson, Kristin A. and Prinz, Fritz B.},\\ntitle={High-throughput screening of inorganic compounds for the\\ndiscovery of novel dielectric and optical materials},\\njournal={Scientific Data},\\nyear={2017},\\nmonth={Jan},\\nday={31},\\npublisher={The Author(s)},\\nvolume={4},\\npages={160134},\\nnote={Data Descriptor},\\nurl={http://dx.doi.org/10.1038/sdata.2016.134}\\n}']\n",
      "File type: json.gz\n",
      "Figshare URL: https://ndownloader.figshare.com/files/13213475\n",
      "SHA256 Hash Digest: 8eb24812148732786cd7c657eccfc6b5ee66533429c2cfbcc4f0059c0295e8b6\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from matminer.datasets import get_all_dataset_info\n",
    "\n",
    "print(get_all_dataset_info(\"dielectric_constant\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "af8befa9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['material_id', 'formula', 'nsites', 'space_group', 'volume',\n",
       "       'structure', 'band_gap', 'e_electronic', 'e_total', 'n',\n",
       "       'poly_electronic', 'poly_total', 'pot_ferroelectric', 'cif', 'meta',\n",
       "       'poscar'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "57279fdf",
   "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>nsites</th>\n",
       "      <th>space_group</th>\n",
       "      <th>volume</th>\n",
       "      <th>band_gap</th>\n",
       "      <th>n</th>\n",
       "      <th>poly_electronic</th>\n",
       "      <th>poly_total</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1056.000000</td>\n",
       "      <td>1056.000000</td>\n",
       "      <td>1056.000000</td>\n",
       "      <td>1056.000000</td>\n",
       "      <td>1056.000000</td>\n",
       "      <td>1056.000000</td>\n",
       "      <td>1056.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>7.530303</td>\n",
       "      <td>142.970644</td>\n",
       "      <td>166.420376</td>\n",
       "      <td>2.119432</td>\n",
       "      <td>2.434886</td>\n",
       "      <td>7.248049</td>\n",
       "      <td>14.777898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>3.388443</td>\n",
       "      <td>67.264591</td>\n",
       "      <td>97.425084</td>\n",
       "      <td>1.604924</td>\n",
       "      <td>1.148849</td>\n",
       "      <td>13.054947</td>\n",
       "      <td>19.435303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>13.980548</td>\n",
       "      <td>0.110000</td>\n",
       "      <td>1.280000</td>\n",
       "      <td>1.630000</td>\n",
       "      <td>2.080000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.000000</td>\n",
       "      <td>82.000000</td>\n",
       "      <td>96.262337</td>\n",
       "      <td>0.890000</td>\n",
       "      <td>1.770000</td>\n",
       "      <td>3.130000</td>\n",
       "      <td>7.557500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>8.000000</td>\n",
       "      <td>163.000000</td>\n",
       "      <td>145.944691</td>\n",
       "      <td>1.730000</td>\n",
       "      <td>2.190000</td>\n",
       "      <td>4.790000</td>\n",
       "      <td>10.540000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>9.000000</td>\n",
       "      <td>194.000000</td>\n",
       "      <td>212.106405</td>\n",
       "      <td>2.885000</td>\n",
       "      <td>2.730000</td>\n",
       "      <td>7.440000</td>\n",
       "      <td>15.482500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>20.000000</td>\n",
       "      <td>229.000000</td>\n",
       "      <td>597.341134</td>\n",
       "      <td>8.320000</td>\n",
       "      <td>16.030000</td>\n",
       "      <td>256.840000</td>\n",
       "      <td>277.780000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            nsites  space_group       volume     band_gap            n  \\\n",
       "count  1056.000000  1056.000000  1056.000000  1056.000000  1056.000000   \n",
       "mean      7.530303   142.970644   166.420376     2.119432     2.434886   \n",
       "std       3.388443    67.264591    97.425084     1.604924     1.148849   \n",
       "min       2.000000     1.000000    13.980548     0.110000     1.280000   \n",
       "25%       5.000000    82.000000    96.262337     0.890000     1.770000   \n",
       "50%       8.000000   163.000000   145.944691     1.730000     2.190000   \n",
       "75%       9.000000   194.000000   212.106405     2.885000     2.730000   \n",
       "max      20.000000   229.000000   597.341134     8.320000    16.030000   \n",
       "\n",
       "       poly_electronic   poly_total  \n",
       "count      1056.000000  1056.000000  \n",
       "mean          7.248049    14.777898  \n",
       "std          13.054947    19.435303  \n",
       "min           1.630000     2.080000  \n",
       "25%           3.130000     7.557500  \n",
       "50%           4.790000    10.540000  \n",
       "75%           7.440000    15.482500  \n",
       "max         256.840000   277.780000  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "0efeebde",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       1.88\n",
       "1       3.52\n",
       "2       1.17\n",
       "3       1.12\n",
       "4       2.87\n",
       "        ... \n",
       "1051    0.87\n",
       "1052    3.60\n",
       "1053    0.14\n",
       "1054    0.21\n",
       "1055    0.26\n",
       "Name: band_gap, Length: 1056, dtype: float64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"band_gap\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "f4c0c354",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "material_id                                                   mp-22881\n",
       "formula                                                          CdCl2\n",
       "nsites                                                               3\n",
       "space_group                                                        166\n",
       "volume                                                       84.298097\n",
       "structure            [[0. 0. 0.] Cd, [ 4.27210959  2.64061969 13.13...\n",
       "band_gap                                                          3.52\n",
       "e_electronic         [[3.34688382, -0.04498543, -0.22379197], [-0.0...\n",
       "e_total              [[7.97018673, -0.29423886, -1.463590159999999]...\n",
       "n                                                                 1.78\n",
       "poly_electronic                                                   3.16\n",
       "poly_total                                                        6.73\n",
       "pot_ferroelectric                                                False\n",
       "cif                  #\\#CIF1.1\\n###################################...\n",
       "meta                 {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...\n",
       "poscar               Cd1 Cl2\\n1.0\\n3.850977 0.072671 5.494462\\n1.78...\n",
       "Name: 1, dtype: object"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.iloc[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "9bd4b9df",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>material_id</th>\n",
       "      <th>formula</th>\n",
       "      <th>nsites</th>\n",
       "      <th>space_group</th>\n",
       "      <th>volume</th>\n",
       "      <th>structure</th>\n",
       "      <th>band_gap</th>\n",
       "      <th>e_electronic</th>\n",
       "      <th>e_total</th>\n",
       "      <th>n</th>\n",
       "      <th>poly_electronic</th>\n",
       "      <th>poly_total</th>\n",
       "      <th>pot_ferroelectric</th>\n",
       "      <th>cif</th>\n",
       "      <th>meta</th>\n",
       "      <th>poscar</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-22881</td>\n",
       "      <td>CdCl2</td>\n",
       "      <td>3</td>\n",
       "      <td>166</td>\n",
       "      <td>84.298097</td>\n",
       "      <td>[[0. 0. 0.] Cd, [ 4.27210959  2.64061969 13.13...</td>\n",
       "      <td>3.52</td>\n",
       "      <td>[[3.34688382, -0.04498543, -0.22379197], [-0.0...</td>\n",
       "      <td>[[7.97018673, -0.29423886, -1.463590159999999]...</td>\n",
       "      <td>1.78</td>\n",
       "      <td>3.16</td>\n",
       "      <td>6.73</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Cd1 Cl2\\n1.0\\n3.850977 0.072671 5.494462\\n1.78...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-560902</td>\n",
       "      <td>MnF2</td>\n",
       "      <td>6</td>\n",
       "      <td>136</td>\n",
       "      <td>82.826401</td>\n",
       "      <td>[[1.677294 2.484476 2.484476] Mn, [0. 0. 0.] M...</td>\n",
       "      <td>2.87</td>\n",
       "      <td>[[2.4239622, 7.452000000000001e-05, 6.06100000...</td>\n",
       "      <td>[[6.44055613, 0.0020446600000000002, 0.0013203...</td>\n",
       "      <td>1.53</td>\n",
       "      <td>2.35</td>\n",
       "      <td>7.12</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLDAUTYPE ...</td>\n",
       "      <td>Mn2 F4\\n1.0\\n3.354588 0.000000 0.000000\\n0.000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>mp-1672</td>\n",
       "      <td>CaS</td>\n",
       "      <td>2</td>\n",
       "      <td>225</td>\n",
       "      <td>46.695124</td>\n",
       "      <td>[[0. 0. 0.] Ca, [2.3336455 1.6501365 4.0419925...</td>\n",
       "      <td>2.38</td>\n",
       "      <td>[[4.98832393, 3.2599999999999997e-06, 1.514e-0...</td>\n",
       "      <td>[[12.09049261, 0.000290789999999, 2.1730000000...</td>\n",
       "      <td>2.23</td>\n",
       "      <td>4.99</td>\n",
       "      <td>12.09</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Ca1 S1\\n1.0\\n3.500468 0.000000 2.020996\\n1.166...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>mp-2530</td>\n",
       "      <td>Li2Te</td>\n",
       "      <td>3</td>\n",
       "      <td>225</td>\n",
       "      <td>69.537895</td>\n",
       "      <td>[[1.33246225 0.942193   2.307892  ] Li, [3.997...</td>\n",
       "      <td>2.50</td>\n",
       "      <td>[[4.96759206, 3.433e-05, -1.184e-05], [3.44600...</td>\n",
       "      <td>[[8.419476580000001, -0.00043652000000000003, ...</td>\n",
       "      <td>2.23</td>\n",
       "      <td>4.97</td>\n",
       "      <td>8.42</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Li2 Te1\\n1.0\\n3.997387 0.000000 2.307892\\n1.33...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>mp-1315</td>\n",
       "      <td>MgS</td>\n",
       "      <td>2</td>\n",
       "      <td>225</td>\n",
       "      <td>35.743489</td>\n",
       "      <td>[[0. 0. 0.] Mg, [2.134732  1.5094835 3.697464 ...</td>\n",
       "      <td>2.76</td>\n",
       "      <td>[[5.4004948200000005, -6.626e-05, -1.568000000...</td>\n",
       "      <td>[[16.01618817, 0.000940069999999, 0.0018529400...</td>\n",
       "      <td>2.32</td>\n",
       "      <td>5.40</td>\n",
       "      <td>16.02</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Mg1 S1\\n1.0\\n3.202098 0.000000 1.848732\\n1.067...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1047</th>\n",
       "      <td>mp-5588</td>\n",
       "      <td>BaSiF6</td>\n",
       "      <td>8</td>\n",
       "      <td>166</td>\n",
       "      <td>110.246829</td>\n",
       "      <td>[[2.003876  2.357052  1.7429385] Ba, [0. 0. 0....</td>\n",
       "      <td>7.42</td>\n",
       "      <td>[[2.18952801, 0.03227471, 0.02386589], [0.0322...</td>\n",
       "      <td>[[8.79266341, -0.39666647, -0.29332863], [-0.3...</td>\n",
       "      <td>1.48</td>\n",
       "      <td>2.19</td>\n",
       "      <td>8.78</td>\n",
       "      <td>True</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Ba1 Si1 F6\\n1.0\\n4.802440 -0.011916 -0.682396\\...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1048</th>\n",
       "      <td>mp-30058</td>\n",
       "      <td>TlSbO3</td>\n",
       "      <td>10</td>\n",
       "      <td>148</td>\n",
       "      <td>196.251592</td>\n",
       "      <td>[[ 4.74259066  2.97725493 13.11556493] Tl, [2....</td>\n",
       "      <td>2.43</td>\n",
       "      <td>[[4.06351653, -0.01514564, -0.06671735], [-0.0...</td>\n",
       "      <td>[[9.88843421, -0.18027358000000002, -0.7942900...</td>\n",
       "      <td>2.00</td>\n",
       "      <td>4.01</td>\n",
       "      <td>9.31</td>\n",
       "      <td>True</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Tl2 Sb2 O6\\n1.0\\n5.229548 0.089115 6.507557\\n2...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1049</th>\n",
       "      <td>mp-23860</td>\n",
       "      <td>SrHCl</td>\n",
       "      <td>6</td>\n",
       "      <td>129</td>\n",
       "      <td>119.914288</td>\n",
       "      <td>[[3.095943   3.095943   5.62892324] Sr, [1.031...</td>\n",
       "      <td>4.04</td>\n",
       "      <td>[[3.51642873, 0.0, 0.0], [0.0, 3.51642873, 0.0...</td>\n",
       "      <td>[[9.40077505, 0.0, 1e-08], [0.0, 9.40075051, 1...</td>\n",
       "      <td>1.87</td>\n",
       "      <td>3.51</td>\n",
       "      <td>9.76</td>\n",
       "      <td>True</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Sr2 H2 Cl2\\n1.0\\n4.127924 0.000000 0.000000\\n0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1050</th>\n",
       "      <td>mp-605034</td>\n",
       "      <td>K2NiF6</td>\n",
       "      <td>9</td>\n",
       "      <td>225</td>\n",
       "      <td>139.856762</td>\n",
       "      <td>[[1.68193375 1.18930675 2.91319475] K, [5.0458...</td>\n",
       "      <td>2.36</td>\n",
       "      <td>[[2.29584983, 1.6e-07, 2.2999999999999997e-07]...</td>\n",
       "      <td>[[5.5624522899999995, -5.730000000000001e-05, ...</td>\n",
       "      <td>1.52</td>\n",
       "      <td>2.30</td>\n",
       "      <td>5.56</td>\n",
       "      <td>True</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLDAUTYPE ...</td>\n",
       "      <td>K2 Ni1 F6\\n1.0\\n5.045801 0.000000 2.913195\\n1....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1052</th>\n",
       "      <td>mp-696944</td>\n",
       "      <td>LaHBr2</td>\n",
       "      <td>8</td>\n",
       "      <td>194</td>\n",
       "      <td>220.041363</td>\n",
       "      <td>[[2.068917   3.58317965 3.70992025] La, [4.400...</td>\n",
       "      <td>3.60</td>\n",
       "      <td>[[4.40504391, 6.1e-07, 0.0], [6.1e-07, 4.40501...</td>\n",
       "      <td>[[8.77136355, 1.649999999999999e-06, 0.0], [1....</td>\n",
       "      <td>2.00</td>\n",
       "      <td>3.99</td>\n",
       "      <td>7.08</td>\n",
       "      <td>True</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>La2 H2 Br4\\n1.0\\n4.137833 0.000000 0.000000\\n-...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>390 rows × 16 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     material_id formula  nsites  space_group      volume  \\\n",
       "1       mp-22881   CdCl2       3          166   84.298097   \n",
       "4      mp-560902    MnF2       6          136   82.826401   \n",
       "11       mp-1672     CaS       2          225   46.695124   \n",
       "12       mp-2530   Li2Te       3          225   69.537895   \n",
       "16       mp-1315     MgS       2          225   35.743489   \n",
       "...          ...     ...     ...          ...         ...   \n",
       "1047     mp-5588  BaSiF6       8          166  110.246829   \n",
       "1048    mp-30058  TlSbO3      10          148  196.251592   \n",
       "1049    mp-23860   SrHCl       6          129  119.914288   \n",
       "1050   mp-605034  K2NiF6       9          225  139.856762   \n",
       "1052   mp-696944  LaHBr2       8          194  220.041363   \n",
       "\n",
       "                                              structure  band_gap  \\\n",
       "1     [[0. 0. 0.] Cd, [ 4.27210959  2.64061969 13.13...      3.52   \n",
       "4     [[1.677294 2.484476 2.484476] Mn, [0. 0. 0.] M...      2.87   \n",
       "11    [[0. 0. 0.] Ca, [2.3336455 1.6501365 4.0419925...      2.38   \n",
       "12    [[1.33246225 0.942193   2.307892  ] Li, [3.997...      2.50   \n",
       "16    [[0. 0. 0.] Mg, [2.134732  1.5094835 3.697464 ...      2.76   \n",
       "...                                                 ...       ...   \n",
       "1047  [[2.003876  2.357052  1.7429385] Ba, [0. 0. 0....      7.42   \n",
       "1048  [[ 4.74259066  2.97725493 13.11556493] Tl, [2....      2.43   \n",
       "1049  [[3.095943   3.095943   5.62892324] Sr, [1.031...      4.04   \n",
       "1050  [[1.68193375 1.18930675 2.91319475] K, [5.0458...      2.36   \n",
       "1052  [[2.068917   3.58317965 3.70992025] La, [4.400...      3.60   \n",
       "\n",
       "                                           e_electronic  \\\n",
       "1     [[3.34688382, -0.04498543, -0.22379197], [-0.0...   \n",
       "4     [[2.4239622, 7.452000000000001e-05, 6.06100000...   \n",
       "11    [[4.98832393, 3.2599999999999997e-06, 1.514e-0...   \n",
       "12    [[4.96759206, 3.433e-05, -1.184e-05], [3.44600...   \n",
       "16    [[5.4004948200000005, -6.626e-05, -1.568000000...   \n",
       "...                                                 ...   \n",
       "1047  [[2.18952801, 0.03227471, 0.02386589], [0.0322...   \n",
       "1048  [[4.06351653, -0.01514564, -0.06671735], [-0.0...   \n",
       "1049  [[3.51642873, 0.0, 0.0], [0.0, 3.51642873, 0.0...   \n",
       "1050  [[2.29584983, 1.6e-07, 2.2999999999999997e-07]...   \n",
       "1052  [[4.40504391, 6.1e-07, 0.0], [6.1e-07, 4.40501...   \n",
       "\n",
       "                                                e_total     n  \\\n",
       "1     [[7.97018673, -0.29423886, -1.463590159999999]...  1.78   \n",
       "4     [[6.44055613, 0.0020446600000000002, 0.0013203...  1.53   \n",
       "11    [[12.09049261, 0.000290789999999, 2.1730000000...  2.23   \n",
       "12    [[8.419476580000001, -0.00043652000000000003, ...  2.23   \n",
       "16    [[16.01618817, 0.000940069999999, 0.0018529400...  2.32   \n",
       "...                                                 ...   ...   \n",
       "1047  [[8.79266341, -0.39666647, -0.29332863], [-0.3...  1.48   \n",
       "1048  [[9.88843421, -0.18027358000000002, -0.7942900...  2.00   \n",
       "1049  [[9.40077505, 0.0, 1e-08], [0.0, 9.40075051, 1...  1.87   \n",
       "1050  [[5.5624522899999995, -5.730000000000001e-05, ...  1.52   \n",
       "1052  [[8.77136355, 1.649999999999999e-06, 0.0], [1....  2.00   \n",
       "\n",
       "      poly_electronic  poly_total  pot_ferroelectric  \\\n",
       "1                3.16        6.73              False   \n",
       "4                2.35        7.12              False   \n",
       "11               4.99       12.09              False   \n",
       "12               4.97        8.42              False   \n",
       "16               5.40       16.02              False   \n",
       "...               ...         ...                ...   \n",
       "1047             2.19        8.78               True   \n",
       "1048             4.01        9.31               True   \n",
       "1049             3.51        9.76               True   \n",
       "1050             2.30        5.56               True   \n",
       "1052             3.99        7.08               True   \n",
       "\n",
       "                                                    cif  \\\n",
       "1     #\\#CIF1.1\\n###################################...   \n",
       "4     #\\#CIF1.1\\n###################################...   \n",
       "11    #\\#CIF1.1\\n###################################...   \n",
       "12    #\\#CIF1.1\\n###################################...   \n",
       "16    #\\#CIF1.1\\n###################################...   \n",
       "...                                                 ...   \n",
       "1047  #\\#CIF1.1\\n###################################...   \n",
       "1048  #\\#CIF1.1\\n###################################...   \n",
       "1049  #\\#CIF1.1\\n###################################...   \n",
       "1050  #\\#CIF1.1\\n###################################...   \n",
       "1052  #\\#CIF1.1\\n###################################...   \n",
       "\n",
       "                                                   meta  \\\n",
       "1     {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "4     {u'incar': u'NELM = 100\\nIBRION = 8\\nLDAUTYPE ...   \n",
       "11    {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "12    {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "16    {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "...                                                 ...   \n",
       "1047  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "1048  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "1049  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "1050  {u'incar': u'NELM = 100\\nIBRION = 8\\nLDAUTYPE ...   \n",
       "1052  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "\n",
       "                                                 poscar  \n",
       "1     Cd1 Cl2\\n1.0\\n3.850977 0.072671 5.494462\\n1.78...  \n",
       "4     Mn2 F4\\n1.0\\n3.354588 0.000000 0.000000\\n0.000...  \n",
       "11    Ca1 S1\\n1.0\\n3.500468 0.000000 2.020996\\n1.166...  \n",
       "12    Li2 Te1\\n1.0\\n3.997387 0.000000 2.307892\\n1.33...  \n",
       "16    Mg1 S1\\n1.0\\n3.202098 0.000000 1.848732\\n1.067...  \n",
       "...                                                 ...  \n",
       "1047  Ba1 Si1 F6\\n1.0\\n4.802440 -0.011916 -0.682396\\...  \n",
       "1048  Tl2 Sb2 O6\\n1.0\\n5.229548 0.089115 6.507557\\n2...  \n",
       "1049  Sr2 H2 Cl2\\n1.0\\n4.127924 0.000000 0.000000\\n0...  \n",
       "1050  K2 Ni1 F6\\n1.0\\n5.045801 0.000000 2.913195\\n1....  \n",
       "1052  La2 H2 Br4\\n1.0\\n4.137833 0.000000 0.000000\\n-...  \n",
       "\n",
       "[390 rows x 16 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Filtre\n",
    "\n",
    "mask = df[\"band_gap\"] > 2.2\n",
    "df[mask]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "445d7fc2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>material_id</th>\n",
       "      <th>formula</th>\n",
       "      <th>volume</th>\n",
       "      <th>structure</th>\n",
       "      <th>band_gap</th>\n",
       "      <th>n</th>\n",
       "      <th>poly_electronic</th>\n",
       "      <th>poly_total</th>\n",
       "      <th>pot_ferroelectric</th>\n",
       "      <th>cif</th>\n",
       "      <th>meta</th>\n",
       "      <th>poscar</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-441</td>\n",
       "      <td>Rb2Te</td>\n",
       "      <td>159.501208</td>\n",
       "      <td>[[1.75725875 1.2425695  3.04366125] Rb, [5.271...</td>\n",
       "      <td>1.88</td>\n",
       "      <td>1.86</td>\n",
       "      <td>3.44</td>\n",
       "      <td>6.23</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Rb2 Te1\\n1.0\\n5.271776 0.000000 3.043661\\n1.75...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-22881</td>\n",
       "      <td>CdCl2</td>\n",
       "      <td>84.298097</td>\n",
       "      <td>[[0. 0. 0.] Cd, [ 4.27210959  2.64061969 13.13...</td>\n",
       "      <td>3.52</td>\n",
       "      <td>1.78</td>\n",
       "      <td>3.16</td>\n",
       "      <td>6.73</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Cd1 Cl2\\n1.0\\n3.850977 0.072671 5.494462\\n1.78...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-28013</td>\n",
       "      <td>MnI2</td>\n",
       "      <td>108.335875</td>\n",
       "      <td>[[0. 0. 0.] Mn, [-2.07904300e-06  2.40067320e+...</td>\n",
       "      <td>1.17</td>\n",
       "      <td>2.23</td>\n",
       "      <td>4.97</td>\n",
       "      <td>10.64</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Mn1 I2\\n1.0\\n4.158086 0.000000 0.000000\\n-2.07...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-567290</td>\n",
       "      <td>LaN</td>\n",
       "      <td>88.162562</td>\n",
       "      <td>[[-1.73309900e-06  2.38611186e+00  5.95256328e...</td>\n",
       "      <td>1.12</td>\n",
       "      <td>2.65</td>\n",
       "      <td>7.04</td>\n",
       "      <td>17.99</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>La2 N2\\n1.0\\n4.132865 0.000000 0.000000\\n-2.06...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-560902</td>\n",
       "      <td>MnF2</td>\n",
       "      <td>82.826401</td>\n",
       "      <td>[[1.677294 2.484476 2.484476] Mn, [0. 0. 0.] M...</td>\n",
       "      <td>2.87</td>\n",
       "      <td>1.53</td>\n",
       "      <td>2.35</td>\n",
       "      <td>7.12</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLDAUTYPE ...</td>\n",
       "      <td>Mn2 F4\\n1.0\\n3.354588 0.000000 0.000000\\n0.000...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id formula      volume  \\\n",
       "0      mp-441   Rb2Te  159.501208   \n",
       "1    mp-22881   CdCl2   84.298097   \n",
       "2    mp-28013    MnI2  108.335875   \n",
       "3   mp-567290     LaN   88.162562   \n",
       "4   mp-560902    MnF2   82.826401   \n",
       "\n",
       "                                           structure  band_gap     n  \\\n",
       "0  [[1.75725875 1.2425695  3.04366125] Rb, [5.271...      1.88  1.86   \n",
       "1  [[0. 0. 0.] Cd, [ 4.27210959  2.64061969 13.13...      3.52  1.78   \n",
       "2  [[0. 0. 0.] Mn, [-2.07904300e-06  2.40067320e+...      1.17  2.23   \n",
       "3  [[-1.73309900e-06  2.38611186e+00  5.95256328e...      1.12  2.65   \n",
       "4  [[1.677294 2.484476 2.484476] Mn, [0. 0. 0.] M...      2.87  1.53   \n",
       "\n",
       "   poly_electronic  poly_total  pot_ferroelectric  \\\n",
       "0             3.44        6.23              False   \n",
       "1             3.16        6.73              False   \n",
       "2             4.97       10.64              False   \n",
       "3             7.04       17.99              False   \n",
       "4             2.35        7.12              False   \n",
       "\n",
       "                                                 cif  \\\n",
       "0  #\\#CIF1.1\\n###################################...   \n",
       "1  #\\#CIF1.1\\n###################################...   \n",
       "2  #\\#CIF1.1\\n###################################...   \n",
       "3  #\\#CIF1.1\\n###################################...   \n",
       "4  #\\#CIF1.1\\n###################################...   \n",
       "\n",
       "                                                meta  \\\n",
       "0  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "1  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "2  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "3  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "4  {u'incar': u'NELM = 100\\nIBRION = 8\\nLDAUTYPE ...   \n",
       "\n",
       "                                              poscar  \n",
       "0  Rb2 Te1\\n1.0\\n5.271776 0.000000 3.043661\\n1.75...  \n",
       "1  Cd1 Cl2\\n1.0\\n3.850977 0.072671 5.494462\\n1.78...  \n",
       "2  Mn1 I2\\n1.0\\n4.158086 0.000000 0.000000\\n-2.07...  \n",
       "3  La2 N2\\n1.0\\n4.132865 0.000000 0.000000\\n-2.06...  \n",
       "4  Mn2 F4\\n1.0\\n3.354588 0.000000 0.000000\\n0.000...  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cleaned_df = df.drop([\"nsites\", \"space_group\", \"e_electronic\", \"e_total\"], axis=1)\n",
    "cleaned_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "c0032f92",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>material_id</th>\n",
       "      <th>formula</th>\n",
       "      <th>nsites</th>\n",
       "      <th>space_group</th>\n",
       "      <th>volume</th>\n",
       "      <th>structure</th>\n",
       "      <th>band_gap</th>\n",
       "      <th>e_electronic</th>\n",
       "      <th>e_total</th>\n",
       "      <th>n</th>\n",
       "      <th>poly_electronic</th>\n",
       "      <th>poly_total</th>\n",
       "      <th>pot_ferroelectric</th>\n",
       "      <th>cif</th>\n",
       "      <th>meta</th>\n",
       "      <th>poscar</th>\n",
       "      <th>poly_ionic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-441</td>\n",
       "      <td>Rb2Te</td>\n",
       "      <td>3</td>\n",
       "      <td>225</td>\n",
       "      <td>159.501208</td>\n",
       "      <td>[[1.75725875 1.2425695  3.04366125] Rb, [5.271...</td>\n",
       "      <td>1.88</td>\n",
       "      <td>[[3.44115795, -3.097e-05, -6.276e-05], [-2.837...</td>\n",
       "      <td>[[6.23414745, -0.00035252, -9.796e-05], [-0.00...</td>\n",
       "      <td>1.86</td>\n",
       "      <td>3.44</td>\n",
       "      <td>6.23</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Rb2 Te1\\n1.0\\n5.271776 0.000000 3.043661\\n1.75...</td>\n",
       "      <td>2.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-22881</td>\n",
       "      <td>CdCl2</td>\n",
       "      <td>3</td>\n",
       "      <td>166</td>\n",
       "      <td>84.298097</td>\n",
       "      <td>[[0. 0. 0.] Cd, [ 4.27210959  2.64061969 13.13...</td>\n",
       "      <td>3.52</td>\n",
       "      <td>[[3.34688382, -0.04498543, -0.22379197], [-0.0...</td>\n",
       "      <td>[[7.97018673, -0.29423886, -1.463590159999999]...</td>\n",
       "      <td>1.78</td>\n",
       "      <td>3.16</td>\n",
       "      <td>6.73</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Cd1 Cl2\\n1.0\\n3.850977 0.072671 5.494462\\n1.78...</td>\n",
       "      <td>3.57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-28013</td>\n",
       "      <td>MnI2</td>\n",
       "      <td>3</td>\n",
       "      <td>164</td>\n",
       "      <td>108.335875</td>\n",
       "      <td>[[0. 0. 0.] Mn, [-2.07904300e-06  2.40067320e+...</td>\n",
       "      <td>1.17</td>\n",
       "      <td>[[5.5430849, -5.28e-06, -2.5030000000000003e-0...</td>\n",
       "      <td>[[13.80606079, 0.0006911900000000001, 9.655e-0...</td>\n",
       "      <td>2.23</td>\n",
       "      <td>4.97</td>\n",
       "      <td>10.64</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>Mn1 I2\\n1.0\\n4.158086 0.000000 0.000000\\n-2.07...</td>\n",
       "      <td>5.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-567290</td>\n",
       "      <td>LaN</td>\n",
       "      <td>4</td>\n",
       "      <td>186</td>\n",
       "      <td>88.162562</td>\n",
       "      <td>[[-1.73309900e-06  2.38611186e+00  5.95256328e...</td>\n",
       "      <td>1.12</td>\n",
       "      <td>[[7.09316738, 7.99e-06, -0.0003864700000000000...</td>\n",
       "      <td>[[16.79535386, 8.199999999999997e-07, -0.00948...</td>\n",
       "      <td>2.65</td>\n",
       "      <td>7.04</td>\n",
       "      <td>17.99</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...</td>\n",
       "      <td>La2 N2\\n1.0\\n4.132865 0.000000 0.000000\\n-2.06...</td>\n",
       "      <td>10.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-560902</td>\n",
       "      <td>MnF2</td>\n",
       "      <td>6</td>\n",
       "      <td>136</td>\n",
       "      <td>82.826401</td>\n",
       "      <td>[[1.677294 2.484476 2.484476] Mn, [0. 0. 0.] M...</td>\n",
       "      <td>2.87</td>\n",
       "      <td>[[2.4239622, 7.452000000000001e-05, 6.06100000...</td>\n",
       "      <td>[[6.44055613, 0.0020446600000000002, 0.0013203...</td>\n",
       "      <td>1.53</td>\n",
       "      <td>2.35</td>\n",
       "      <td>7.12</td>\n",
       "      <td>False</td>\n",
       "      <td>#\\#CIF1.1\\n###################################...</td>\n",
       "      <td>{u'incar': u'NELM = 100\\nIBRION = 8\\nLDAUTYPE ...</td>\n",
       "      <td>Mn2 F4\\n1.0\\n3.354588 0.000000 0.000000\\n0.000...</td>\n",
       "      <td>4.77</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  material_id formula  nsites  space_group      volume  \\\n",
       "0      mp-441   Rb2Te       3          225  159.501208   \n",
       "1    mp-22881   CdCl2       3          166   84.298097   \n",
       "2    mp-28013    MnI2       3          164  108.335875   \n",
       "3   mp-567290     LaN       4          186   88.162562   \n",
       "4   mp-560902    MnF2       6          136   82.826401   \n",
       "\n",
       "                                           structure  band_gap  \\\n",
       "0  [[1.75725875 1.2425695  3.04366125] Rb, [5.271...      1.88   \n",
       "1  [[0. 0. 0.] Cd, [ 4.27210959  2.64061969 13.13...      3.52   \n",
       "2  [[0. 0. 0.] Mn, [-2.07904300e-06  2.40067320e+...      1.17   \n",
       "3  [[-1.73309900e-06  2.38611186e+00  5.95256328e...      1.12   \n",
       "4  [[1.677294 2.484476 2.484476] Mn, [0. 0. 0.] M...      2.87   \n",
       "\n",
       "                                        e_electronic  \\\n",
       "0  [[3.44115795, -3.097e-05, -6.276e-05], [-2.837...   \n",
       "1  [[3.34688382, -0.04498543, -0.22379197], [-0.0...   \n",
       "2  [[5.5430849, -5.28e-06, -2.5030000000000003e-0...   \n",
       "3  [[7.09316738, 7.99e-06, -0.0003864700000000000...   \n",
       "4  [[2.4239622, 7.452000000000001e-05, 6.06100000...   \n",
       "\n",
       "                                             e_total     n  poly_electronic  \\\n",
       "0  [[6.23414745, -0.00035252, -9.796e-05], [-0.00...  1.86             3.44   \n",
       "1  [[7.97018673, -0.29423886, -1.463590159999999]...  1.78             3.16   \n",
       "2  [[13.80606079, 0.0006911900000000001, 9.655e-0...  2.23             4.97   \n",
       "3  [[16.79535386, 8.199999999999997e-07, -0.00948...  2.65             7.04   \n",
       "4  [[6.44055613, 0.0020446600000000002, 0.0013203...  1.53             2.35   \n",
       "\n",
       "   poly_total  pot_ferroelectric  \\\n",
       "0        6.23              False   \n",
       "1        6.73              False   \n",
       "2       10.64              False   \n",
       "3       17.99              False   \n",
       "4        7.12              False   \n",
       "\n",
       "                                                 cif  \\\n",
       "0  #\\#CIF1.1\\n###################################...   \n",
       "1  #\\#CIF1.1\\n###################################...   \n",
       "2  #\\#CIF1.1\\n###################################...   \n",
       "3  #\\#CIF1.1\\n###################################...   \n",
       "4  #\\#CIF1.1\\n###################################...   \n",
       "\n",
       "                                                meta  \\\n",
       "0  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "1  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "2  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "3  {u'incar': u'NELM = 100\\nIBRION = 8\\nLWAVE = F...   \n",
       "4  {u'incar': u'NELM = 100\\nIBRION = 8\\nLDAUTYPE ...   \n",
       "\n",
       "                                              poscar  poly_ionic  \n",
       "0  Rb2 Te1\\n1.0\\n5.271776 0.000000 3.043661\\n1.75...        2.79  \n",
       "1  Cd1 Cl2\\n1.0\\n3.850977 0.072671 5.494462\\n1.78...        3.57  \n",
       "2  Mn1 I2\\n1.0\\n4.158086 0.000000 0.000000\\n-2.07...        5.67  \n",
       "3  La2 N2\\n1.0\\n4.132865 0.000000 0.000000\\n-2.06...       10.95  \n",
       "4  Mn2 F4\\n1.0\\n3.354588 0.000000 0.000000\\n0.000...        4.77  "
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"poly_ionic\"] = df[\"poly_total\"] - df[\"poly_electronic\"]\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23a986be",
   "metadata": {},
   "source": [
    "### Featurisers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "5e32b965",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pymatgen.core import Composition\n",
    "from matminer.featurizers.composition import ElementFraction\n",
    "\n",
    "fe2o3 = Composition(\"Fe2O3\")\n",
    "ef = ElementFraction()\n",
    "\n",
    "ef_features = ef.featurize(fe2o3)\n",
    "ef_feature_labels = ef.feature_labels()\n",
    "# ef_feature_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "e2c11883",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6 O\n",
      "0.4 Fe\n"
     ]
    }
   ],
   "source": [
    "print(ef_features[7], ef_feature_labels[7]) # the composition of Fe2O3, oxygen is 3/5 = 0.6\n",
    "print(ef_features[25], ef_feature_labels[25]) # the composition of Fe2O3, oxygen is 2/5 = 0.4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "a266a87c",
   "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>formula</th>\n",
       "      <th>bulk_modulus</th>\n",
       "      <th>shear_modulus</th>\n",
       "      <th>composition</th>\n",
       "      <th>material_id</th>\n",
       "      <th>structure</th>\n",
       "      <th>brgoch_feats</th>\n",
       "      <th>suspect_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AlPt3</td>\n",
       "      <td>225.230461</td>\n",
       "      <td>91.197748</td>\n",
       "      <td>(Al, Pt)</td>\n",
       "      <td>mp-188</td>\n",
       "      <td>[[0. 0. 0.] Al, [0.         1.96140395 1.96140...</td>\n",
       "      <td>{'atomic_number_feat_1': 123.5, 'atomic_number...</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Mn2Nb</td>\n",
       "      <td>232.696340</td>\n",
       "      <td>74.590157</td>\n",
       "      <td>(Mn, Nb)</td>\n",
       "      <td>mp-12659</td>\n",
       "      <td>[[-2.23765223e-08  1.42974191e+00  5.92614104e...</td>\n",
       "      <td>{'atomic_number_feat_1': 45.5, 'atomic_number_...</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>HfO2</td>\n",
       "      <td>204.573433</td>\n",
       "      <td>98.564374</td>\n",
       "      <td>(Hf, O)</td>\n",
       "      <td>mp-352</td>\n",
       "      <td>[[2.24450185 3.85793022 4.83390736] O, [2.7788...</td>\n",
       "      <td>{'atomic_number_feat_1': 44.0, 'atomic_number_...</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cu3Pt</td>\n",
       "      <td>159.312640</td>\n",
       "      <td>51.778816</td>\n",
       "      <td>(Cu, Pt)</td>\n",
       "      <td>mp-12086</td>\n",
       "      <td>[[0.         1.86144248 1.86144248] Cu, [1.861...</td>\n",
       "      <td>{'atomic_number_feat_1': 82.5, 'atomic_number_...</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Mg3Pt</td>\n",
       "      <td>69.637565</td>\n",
       "      <td>27.588765</td>\n",
       "      <td>(Mg, Pt)</td>\n",
       "      <td>mp-18707</td>\n",
       "      <td>[[0.         0.         2.73626461] Mg, [0.   ...</td>\n",
       "      <td>{'atomic_number_feat_1': 57.0, 'atomic_number_...</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  formula  bulk_modulus  shear_modulus composition material_id  \\\n",
       "0   AlPt3    225.230461      91.197748    (Al, Pt)      mp-188   \n",
       "1   Mn2Nb    232.696340      74.590157    (Mn, Nb)    mp-12659   \n",
       "2    HfO2    204.573433      98.564374     (Hf, O)      mp-352   \n",
       "3   Cu3Pt    159.312640      51.778816    (Cu, Pt)    mp-12086   \n",
       "4   Mg3Pt     69.637565      27.588765    (Mg, Pt)    mp-18707   \n",
       "\n",
       "                                           structure  \\\n",
       "0  [[0. 0. 0.] Al, [0.         1.96140395 1.96140...   \n",
       "1  [[-2.23765223e-08  1.42974191e+00  5.92614104e...   \n",
       "2  [[2.24450185 3.85793022 4.83390736] O, [2.7788...   \n",
       "3  [[0.         1.86144248 1.86144248] Cu, [1.861...   \n",
       "4  [[0.         0.         2.73626461] Mg, [0.   ...   \n",
       "\n",
       "                                        brgoch_feats  suspect_value  \n",
       "0  {'atomic_number_feat_1': 123.5, 'atomic_number...          False  \n",
       "1  {'atomic_number_feat_1': 45.5, 'atomic_number_...          False  \n",
       "2  {'atomic_number_feat_1': 44.0, 'atomic_number_...          False  \n",
       "3  {'atomic_number_feat_1': 82.5, 'atomic_number_...          False  \n",
       "4  {'atomic_number_feat_1': 57.0, 'atomic_number_...          False  "
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from matminer.datasets.dataset_retrieval import load_dataset\n",
    "\n",
    "df = load_dataset(\"brgoch_superhard_training\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "ac56e234",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 2574 entries, 0 to 2573\n",
      "Data columns (total 8 columns):\n",
      " #   Column         Non-Null Count  Dtype  \n",
      "---  ------         --------------  -----  \n",
      " 0   formula        2574 non-null   object \n",
      " 1   bulk_modulus   2574 non-null   float64\n",
      " 2   shear_modulus  2574 non-null   float64\n",
      " 3   composition    2574 non-null   object \n",
      " 4   material_id    2540 non-null   object \n",
      " 5   structure      2540 non-null   object \n",
      " 6   brgoch_feats   2574 non-null   object \n",
      " 7   suspect_value  2574 non-null   bool   \n",
      "dtypes: bool(1), float64(2), object(5)\n",
      "memory usage: 163.4+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "77f36cdb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ba1a303150f2438abf0a179204d7cd67",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "ElementFraction:   0%|          | 0/2574 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>formula</th>\n",
       "      <th>bulk_modulus</th>\n",
       "      <th>shear_modulus</th>\n",
       "      <th>composition</th>\n",
       "      <th>material_id</th>\n",
       "      <th>structure</th>\n",
       "      <th>brgoch_feats</th>\n",
       "      <th>suspect_value</th>\n",
       "      <th>H</th>\n",
       "      <th>He</th>\n",
       "      <th>...</th>\n",
       "      <th>Pu</th>\n",
       "      <th>Am</th>\n",
       "      <th>Cm</th>\n",
       "      <th>Bk</th>\n",
       "      <th>Cf</th>\n",
       "      <th>Es</th>\n",
       "      <th>Fm</th>\n",
       "      <th>Md</th>\n",
       "      <th>No</th>\n",
       "      <th>Lr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AlPt3</td>\n",
       "      <td>225.230461</td>\n",
       "      <td>91.197748</td>\n",
       "      <td>(Al, Pt)</td>\n",
       "      <td>mp-188</td>\n",
       "      <td>[[0. 0. 0.] Al, [0.         1.96140395 1.96140...</td>\n",
       "      <td>{'atomic_number_feat_1': 123.5, 'atomic_number...</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Mn2Nb</td>\n",
       "      <td>232.696340</td>\n",
       "      <td>74.590157</td>\n",
       "      <td>(Mn, Nb)</td>\n",
       "      <td>mp-12659</td>\n",
       "      <td>[[-2.23765223e-08  1.42974191e+00  5.92614104e...</td>\n",
       "      <td>{'atomic_number_feat_1': 45.5, 'atomic_number_...</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>HfO2</td>\n",
       "      <td>204.573433</td>\n",
       "      <td>98.564374</td>\n",
       "      <td>(Hf, O)</td>\n",
       "      <td>mp-352</td>\n",
       "      <td>[[2.24450185 3.85793022 4.83390736] O, [2.7788...</td>\n",
       "      <td>{'atomic_number_feat_1': 44.0, 'atomic_number_...</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cu3Pt</td>\n",
       "      <td>159.312640</td>\n",
       "      <td>51.778816</td>\n",
       "      <td>(Cu, Pt)</td>\n",
       "      <td>mp-12086</td>\n",
       "      <td>[[0.         1.86144248 1.86144248] Cu, [1.861...</td>\n",
       "      <td>{'atomic_number_feat_1': 82.5, 'atomic_number_...</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Mg3Pt</td>\n",
       "      <td>69.637565</td>\n",
       "      <td>27.588765</td>\n",
       "      <td>(Mg, Pt)</td>\n",
       "      <td>mp-18707</td>\n",
       "      <td>[[0.         0.         2.73626461] Mg, [0.   ...</td>\n",
       "      <td>{'atomic_number_feat_1': 57.0, 'atomic_number_...</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2569</th>\n",
       "      <td>MgAsPt5</td>\n",
       "      <td>181.857661</td>\n",
       "      <td>70.362793</td>\n",
       "      <td>(Mg, As, Pt)</td>\n",
       "      <td>mp-865146</td>\n",
       "      <td>[[1.99379423 1.99379423 0.        ] Mg, [0.   ...</td>\n",
       "      <td>{'atomic_number_feat_1': 145.0, 'atomic_number...</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2570</th>\n",
       "      <td>YCBr</td>\n",
       "      <td>22.970769</td>\n",
       "      <td>9.733656</td>\n",
       "      <td>(Y, C, Br)</td>\n",
       "      <td>mp-643367</td>\n",
       "      <td>[[0.66584097 1.92338676 8.4823332 ] Y, [-0.645...</td>\n",
       "      <td>{'atomic_number_feat_1': 26.6666666666667, 'at...</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2571</th>\n",
       "      <td>NaTiO2</td>\n",
       "      <td>98.358957</td>\n",
       "      <td>70.565034</td>\n",
       "      <td>(Na, Ti, O)</td>\n",
       "      <td>mp-7002</td>\n",
       "      <td>[[ 3.31024633  2.00832854 11.99124842] O, [1.0...</td>\n",
       "      <td>{'atomic_number_feat_1': 16.3333333333333, 'at...</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2572</th>\n",
       "      <td>KCuSe</td>\n",
       "      <td>34.111565</td>\n",
       "      <td>21.586576</td>\n",
       "      <td>(K, Cu, Se)</td>\n",
       "      <td>mp-7435</td>\n",
       "      <td>[[0. 0. 0.] K, [0.         0.         4.903557...</td>\n",
       "      <td>{'atomic_number_feat_1': 27.3333333333333, 'at...</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2573</th>\n",
       "      <td>Y3Al5O12</td>\n",
       "      <td>168.580800</td>\n",
       "      <td>101.413900</td>\n",
       "      <td>(Y, Al, O)</td>\n",
       "      <td>mp-3050</td>\n",
       "      <td>[[ 3.55901406  5.8271786  -3.97983702] O, [ 5....</td>\n",
       "      <td>{'atomic_number_feat_1': 92.6666666666667, 'at...</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2574 rows × 111 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       formula  bulk_modulus  shear_modulus   composition material_id  \\\n",
       "0        AlPt3    225.230461      91.197748      (Al, Pt)      mp-188   \n",
       "1        Mn2Nb    232.696340      74.590157      (Mn, Nb)    mp-12659   \n",
       "2         HfO2    204.573433      98.564374       (Hf, O)      mp-352   \n",
       "3        Cu3Pt    159.312640      51.778816      (Cu, Pt)    mp-12086   \n",
       "4        Mg3Pt     69.637565      27.588765      (Mg, Pt)    mp-18707   \n",
       "...        ...           ...            ...           ...         ...   \n",
       "2569   MgAsPt5    181.857661      70.362793  (Mg, As, Pt)   mp-865146   \n",
       "2570      YCBr     22.970769       9.733656    (Y, C, Br)   mp-643367   \n",
       "2571    NaTiO2     98.358957      70.565034   (Na, Ti, O)     mp-7002   \n",
       "2572     KCuSe     34.111565      21.586576   (K, Cu, Se)     mp-7435   \n",
       "2573  Y3Al5O12    168.580800     101.413900    (Y, Al, O)     mp-3050   \n",
       "\n",
       "                                              structure  \\\n",
       "0     [[0. 0. 0.] Al, [0.         1.96140395 1.96140...   \n",
       "1     [[-2.23765223e-08  1.42974191e+00  5.92614104e...   \n",
       "2     [[2.24450185 3.85793022 4.83390736] O, [2.7788...   \n",
       "3     [[0.         1.86144248 1.86144248] Cu, [1.861...   \n",
       "4     [[0.         0.         2.73626461] Mg, [0.   ...   \n",
       "...                                                 ...   \n",
       "2569  [[1.99379423 1.99379423 0.        ] Mg, [0.   ...   \n",
       "2570  [[0.66584097 1.92338676 8.4823332 ] Y, [-0.645...   \n",
       "2571  [[ 3.31024633  2.00832854 11.99124842] O, [1.0...   \n",
       "2572  [[0. 0. 0.] K, [0.         0.         4.903557...   \n",
       "2573  [[ 3.55901406  5.8271786  -3.97983702] O, [ 5....   \n",
       "\n",
       "                                           brgoch_feats  suspect_value  H  He  \\\n",
       "0     {'atomic_number_feat_1': 123.5, 'atomic_number...          False  0   0   \n",
       "1     {'atomic_number_feat_1': 45.5, 'atomic_number_...          False  0   0   \n",
       "2     {'atomic_number_feat_1': 44.0, 'atomic_number_...          False  0   0   \n",
       "3     {'atomic_number_feat_1': 82.5, 'atomic_number_...          False  0   0   \n",
       "4     {'atomic_number_feat_1': 57.0, 'atomic_number_...          False  0   0   \n",
       "...                                                 ...            ... ..  ..   \n",
       "2569  {'atomic_number_feat_1': 145.0, 'atomic_number...          False  0   0   \n",
       "2570  {'atomic_number_feat_1': 26.6666666666667, 'at...          False  0   0   \n",
       "2571  {'atomic_number_feat_1': 16.3333333333333, 'at...          False  0   0   \n",
       "2572  {'atomic_number_feat_1': 27.3333333333333, 'at...          False  0   0   \n",
       "2573  {'atomic_number_feat_1': 92.6666666666667, 'at...          False  0   0   \n",
       "\n",
       "      ...  Pu  Am  Cm  Bk  Cf  Es  Fm  Md  No  Lr  \n",
       "0     ...   0   0   0   0   0   0   0   0   0   0  \n",
       "1     ...   0   0   0   0   0   0   0   0   0   0  \n",
       "2     ...   0   0   0   0   0   0   0   0   0   0  \n",
       "3     ...   0   0   0   0   0   0   0   0   0   0  \n",
       "4     ...   0   0   0   0   0   0   0   0   0   0  \n",
       "...   ...  ..  ..  ..  ..  ..  ..  ..  ..  ..  ..  \n",
       "2569  ...   0   0   0   0   0   0   0   0   0   0  \n",
       "2570  ...   0   0   0   0   0   0   0   0   0   0  \n",
       "2571  ...   0   0   0   0   0   0   0   0   0   0  \n",
       "2572  ...   0   0   0   0   0   0   0   0   0   0  \n",
       "2573  ...   0   0   0   0   0   0   0   0   0   0  \n",
       "\n",
       "[2574 rows x 111 columns]"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ef.featurize_dataframe(df, \"composition\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6cb1b4d4",
   "metadata": {},
   "source": [
    "### Structural featuriser"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "aae67b3e",
   "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>mpid</th>\n",
       "      <th>eps_electronic</th>\n",
       "      <th>eps_total</th>\n",
       "      <th>last phdos peak</th>\n",
       "      <th>structure</th>\n",
       "      <th>formula</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-1000</td>\n",
       "      <td>6.311555</td>\n",
       "      <td>12.773454</td>\n",
       "      <td>98.585771</td>\n",
       "      <td>[[2.8943817  2.04663693 5.01321616] Te, [0. 0....</td>\n",
       "      <td>BaTe</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-1002124</td>\n",
       "      <td>24.137743</td>\n",
       "      <td>32.965593</td>\n",
       "      <td>677.585725</td>\n",
       "      <td>[[0. 0. 0.] Hf, [-3.78195772 -3.78195772 -3.78...</td>\n",
       "      <td>HfC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-1002164</td>\n",
       "      <td>8.111021</td>\n",
       "      <td>11.169464</td>\n",
       "      <td>761.585719</td>\n",
       "      <td>[[0. 0. 0.] Ge, [ 3.45311592  3.45311592 -3.45...</td>\n",
       "      <td>GeC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-10044</td>\n",
       "      <td>10.032168</td>\n",
       "      <td>10.128936</td>\n",
       "      <td>701.585723</td>\n",
       "      <td>[[0.98372595 0.69559929 1.70386332] B, [0. 0. ...</td>\n",
       "      <td>BAs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-1008223</td>\n",
       "      <td>3.979201</td>\n",
       "      <td>6.394043</td>\n",
       "      <td>204.585763</td>\n",
       "      <td>[[0. 0. 0.] Ca, [ 4.95  4.95 -4.95] Se]</td>\n",
       "      <td>CaSe</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         mpid  eps_electronic  eps_total  last phdos peak  \\\n",
       "0     mp-1000        6.311555  12.773454        98.585771   \n",
       "1  mp-1002124       24.137743  32.965593       677.585725   \n",
       "2  mp-1002164        8.111021  11.169464       761.585719   \n",
       "3    mp-10044       10.032168  10.128936       701.585723   \n",
       "4  mp-1008223        3.979201   6.394043       204.585763   \n",
       "\n",
       "                                           structure formula  \n",
       "0  [[2.8943817  2.04663693 5.01321616] Te, [0. 0....    BaTe  \n",
       "1  [[0. 0. 0.] Hf, [-3.78195772 -3.78195772 -3.78...     HfC  \n",
       "2  [[0. 0. 0.] Ge, [ 3.45311592  3.45311592 -3.45...     GeC  \n",
       "3  [[0.98372595 0.69559929 1.70386332] B, [0. 0. ...     BAs  \n",
       "4            [[0. 0. 0.] Ca, [ 4.95  4.95 -4.95] Se]    CaSe  "
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = load_dataset(\"phonon_dielectric_mp\")\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "0f94745d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6c136721a8534e5bb6555c51b63a7963",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "DensityFeatures:   0%|          | 0/1296 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mpid</th>\n",
       "      <th>eps_electronic</th>\n",
       "      <th>eps_total</th>\n",
       "      <th>last phdos peak</th>\n",
       "      <th>structure</th>\n",
       "      <th>formula</th>\n",
       "      <th>density</th>\n",
       "      <th>vpa</th>\n",
       "      <th>packing fraction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-1000</td>\n",
       "      <td>6.311555</td>\n",
       "      <td>12.773454</td>\n",
       "      <td>98.585771</td>\n",
       "      <td>[[2.8943817  2.04663693 5.01321616] Te, [0. 0....</td>\n",
       "      <td>BaTe</td>\n",
       "      <td>4.937886</td>\n",
       "      <td>44.545547</td>\n",
       "      <td>0.596286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-1002124</td>\n",
       "      <td>24.137743</td>\n",
       "      <td>32.965593</td>\n",
       "      <td>677.585725</td>\n",
       "      <td>[[0. 0. 0.] Hf, [-3.78195772 -3.78195772 -3.78...</td>\n",
       "      <td>HfC</td>\n",
       "      <td>9.868234</td>\n",
       "      <td>16.027886</td>\n",
       "      <td>0.531426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-1002164</td>\n",
       "      <td>8.111021</td>\n",
       "      <td>11.169464</td>\n",
       "      <td>761.585719</td>\n",
       "      <td>[[0. 0. 0.] Ge, [ 3.45311592  3.45311592 -3.45...</td>\n",
       "      <td>GeC</td>\n",
       "      <td>5.760895</td>\n",
       "      <td>12.199996</td>\n",
       "      <td>0.394180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-10044</td>\n",
       "      <td>10.032168</td>\n",
       "      <td>10.128936</td>\n",
       "      <td>701.585723</td>\n",
       "      <td>[[0.98372595 0.69559929 1.70386332] B, [0. 0. ...</td>\n",
       "      <td>BAs</td>\n",
       "      <td>5.087634</td>\n",
       "      <td>13.991016</td>\n",
       "      <td>0.319600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-1008223</td>\n",
       "      <td>3.979201</td>\n",
       "      <td>6.394043</td>\n",
       "      <td>204.585763</td>\n",
       "      <td>[[0. 0. 0.] Ca, [ 4.95  4.95 -4.95] Se]</td>\n",
       "      <td>CaSe</td>\n",
       "      <td>2.750191</td>\n",
       "      <td>35.937000</td>\n",
       "      <td>0.428523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1291</th>\n",
       "      <td>mp-998603</td>\n",
       "      <td>4.178159</td>\n",
       "      <td>14.155681</td>\n",
       "      <td>142.585768</td>\n",
       "      <td>[[1.69099645e-03 3.81913207e+00 1.07685858e-01...</td>\n",
       "      <td>RbPbBr3</td>\n",
       "      <td>4.581318</td>\n",
       "      <td>38.593148</td>\n",
       "      <td>0.507357</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1292</th>\n",
       "      <td>mp-998604</td>\n",
       "      <td>3.548202</td>\n",
       "      <td>13.938313</td>\n",
       "      <td>223.585761</td>\n",
       "      <td>[[1.64439731e-03 3.64832409e+00 1.03287739e-01...</td>\n",
       "      <td>RbPbCl3</td>\n",
       "      <td>3.933733</td>\n",
       "      <td>33.688084</td>\n",
       "      <td>0.542370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1293</th>\n",
       "      <td>mp-998612</td>\n",
       "      <td>3.960980</td>\n",
       "      <td>9.617663</td>\n",
       "      <td>219.718383</td>\n",
       "      <td>[[-3.66731982 -1.91142875  2.96640499] K, [ 3....</td>\n",
       "      <td>KGeBr3</td>\n",
       "      <td>3.151275</td>\n",
       "      <td>37.038786</td>\n",
       "      <td>0.388217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1294</th>\n",
       "      <td>mp-999498</td>\n",
       "      <td>4.613954</td>\n",
       "      <td>4.972619</td>\n",
       "      <td>1090.585692</td>\n",
       "      <td>[[ 1.57631457 -0.32583322 -1.57631457] N, [ 1....</td>\n",
       "      <td>N2</td>\n",
       "      <td>3.379498</td>\n",
       "      <td>6.882287</td>\n",
       "      <td>0.167146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1295</th>\n",
       "      <td>mp-999549</td>\n",
       "      <td>13.421260</td>\n",
       "      <td>13.456055</td>\n",
       "      <td>1038.585697</td>\n",
       "      <td>[[1.46542555 0.84617794 1.95957081] W, [0.0000...</td>\n",
       "      <td>WN2</td>\n",
       "      <td>12.066238</td>\n",
       "      <td>9.718324</td>\n",
       "      <td>0.432403</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1296 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            mpid  eps_electronic  eps_total  last phdos peak  \\\n",
       "0        mp-1000        6.311555  12.773454        98.585771   \n",
       "1     mp-1002124       24.137743  32.965593       677.585725   \n",
       "2     mp-1002164        8.111021  11.169464       761.585719   \n",
       "3       mp-10044       10.032168  10.128936       701.585723   \n",
       "4     mp-1008223        3.979201   6.394043       204.585763   \n",
       "...          ...             ...        ...              ...   \n",
       "1291   mp-998603        4.178159  14.155681       142.585768   \n",
       "1292   mp-998604        3.548202  13.938313       223.585761   \n",
       "1293   mp-998612        3.960980   9.617663       219.718383   \n",
       "1294   mp-999498        4.613954   4.972619      1090.585692   \n",
       "1295   mp-999549       13.421260  13.456055      1038.585697   \n",
       "\n",
       "                                              structure  formula    density  \\\n",
       "0     [[2.8943817  2.04663693 5.01321616] Te, [0. 0....     BaTe   4.937886   \n",
       "1     [[0. 0. 0.] Hf, [-3.78195772 -3.78195772 -3.78...      HfC   9.868234   \n",
       "2     [[0. 0. 0.] Ge, [ 3.45311592  3.45311592 -3.45...      GeC   5.760895   \n",
       "3     [[0.98372595 0.69559929 1.70386332] B, [0. 0. ...      BAs   5.087634   \n",
       "4               [[0. 0. 0.] Ca, [ 4.95  4.95 -4.95] Se]     CaSe   2.750191   \n",
       "...                                                 ...      ...        ...   \n",
       "1291  [[1.69099645e-03 3.81913207e+00 1.07685858e-01...  RbPbBr3   4.581318   \n",
       "1292  [[1.64439731e-03 3.64832409e+00 1.03287739e-01...  RbPbCl3   3.933733   \n",
       "1293  [[-3.66731982 -1.91142875  2.96640499] K, [ 3....   KGeBr3   3.151275   \n",
       "1294  [[ 1.57631457 -0.32583322 -1.57631457] N, [ 1....       N2   3.379498   \n",
       "1295  [[1.46542555 0.84617794 1.95957081] W, [0.0000...      WN2  12.066238   \n",
       "\n",
       "            vpa  packing fraction  \n",
       "0     44.545547          0.596286  \n",
       "1     16.027886          0.531426  \n",
       "2     12.199996          0.394180  \n",
       "3     13.991016          0.319600  \n",
       "4     35.937000          0.428523  \n",
       "...         ...               ...  \n",
       "1291  38.593148          0.507357  \n",
       "1292  33.688084          0.542370  \n",
       "1293  37.038786          0.388217  \n",
       "1294   6.882287          0.167146  \n",
       "1295   9.718324          0.432403  \n",
       "\n",
       "[1296 rows x 9 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from matminer.featurizers.structure import DensityFeatures\n",
    "\n",
    "density = DensityFeatures()\n",
    "density.featurize_dataframe(df, \"structure\")  # new three columns are added: density, volum/atom, packing fraction"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "41455dec",
   "metadata": {},
   "source": [
    "## Conversion Featurisers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "24e55902",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "df069668bdfb46caa365d02a79189beb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "StrToComposition:   0%|          | 0/1296 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mpid</th>\n",
       "      <th>eps_electronic</th>\n",
       "      <th>eps_total</th>\n",
       "      <th>last phdos peak</th>\n",
       "      <th>structure</th>\n",
       "      <th>formula</th>\n",
       "      <th>composition</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mp-1000</td>\n",
       "      <td>6.311555</td>\n",
       "      <td>12.773454</td>\n",
       "      <td>98.585771</td>\n",
       "      <td>[[2.8943817  2.04663693 5.01321616] Te, [0. 0....</td>\n",
       "      <td>BaTe</td>\n",
       "      <td>(Ba, Te)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mp-1002124</td>\n",
       "      <td>24.137743</td>\n",
       "      <td>32.965593</td>\n",
       "      <td>677.585725</td>\n",
       "      <td>[[0. 0. 0.] Hf, [-3.78195772 -3.78195772 -3.78...</td>\n",
       "      <td>HfC</td>\n",
       "      <td>(Hf, C)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mp-1002164</td>\n",
       "      <td>8.111021</td>\n",
       "      <td>11.169464</td>\n",
       "      <td>761.585719</td>\n",
       "      <td>[[0. 0. 0.] Ge, [ 3.45311592  3.45311592 -3.45...</td>\n",
       "      <td>GeC</td>\n",
       "      <td>(Ge, C)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mp-10044</td>\n",
       "      <td>10.032168</td>\n",
       "      <td>10.128936</td>\n",
       "      <td>701.585723</td>\n",
       "      <td>[[0.98372595 0.69559929 1.70386332] B, [0. 0. ...</td>\n",
       "      <td>BAs</td>\n",
       "      <td>(B, As)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mp-1008223</td>\n",
       "      <td>3.979201</td>\n",
       "      <td>6.394043</td>\n",
       "      <td>204.585763</td>\n",
       "      <td>[[0. 0. 0.] Ca, [ 4.95  4.95 -4.95] Se]</td>\n",
       "      <td>CaSe</td>\n",
       "      <td>(Ca, Se)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1291</th>\n",
       "      <td>mp-998603</td>\n",
       "      <td>4.178159</td>\n",
       "      <td>14.155681</td>\n",
       "      <td>142.585768</td>\n",
       "      <td>[[1.69099645e-03 3.81913207e+00 1.07685858e-01...</td>\n",
       "      <td>RbPbBr3</td>\n",
       "      <td>(Rb, Pb, Br)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1292</th>\n",
       "      <td>mp-998604</td>\n",
       "      <td>3.548202</td>\n",
       "      <td>13.938313</td>\n",
       "      <td>223.585761</td>\n",
       "      <td>[[1.64439731e-03 3.64832409e+00 1.03287739e-01...</td>\n",
       "      <td>RbPbCl3</td>\n",
       "      <td>(Rb, Pb, Cl)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1293</th>\n",
       "      <td>mp-998612</td>\n",
       "      <td>3.960980</td>\n",
       "      <td>9.617663</td>\n",
       "      <td>219.718383</td>\n",
       "      <td>[[-3.66731982 -1.91142875  2.96640499] K, [ 3....</td>\n",
       "      <td>KGeBr3</td>\n",
       "      <td>(K, Ge, Br)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1294</th>\n",
       "      <td>mp-999498</td>\n",
       "      <td>4.613954</td>\n",
       "      <td>4.972619</td>\n",
       "      <td>1090.585692</td>\n",
       "      <td>[[ 1.57631457 -0.32583322 -1.57631457] N, [ 1....</td>\n",
       "      <td>N2</td>\n",
       "      <td>(N)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1295</th>\n",
       "      <td>mp-999549</td>\n",
       "      <td>13.421260</td>\n",
       "      <td>13.456055</td>\n",
       "      <td>1038.585697</td>\n",
       "      <td>[[1.46542555 0.84617794 1.95957081] W, [0.0000...</td>\n",
       "      <td>WN2</td>\n",
       "      <td>(W, N)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1296 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            mpid  eps_electronic  eps_total  last phdos peak  \\\n",
       "0        mp-1000        6.311555  12.773454        98.585771   \n",
       "1     mp-1002124       24.137743  32.965593       677.585725   \n",
       "2     mp-1002164        8.111021  11.169464       761.585719   \n",
       "3       mp-10044       10.032168  10.128936       701.585723   \n",
       "4     mp-1008223        3.979201   6.394043       204.585763   \n",
       "...          ...             ...        ...              ...   \n",
       "1291   mp-998603        4.178159  14.155681       142.585768   \n",
       "1292   mp-998604        3.548202  13.938313       223.585761   \n",
       "1293   mp-998612        3.960980   9.617663       219.718383   \n",
       "1294   mp-999498        4.613954   4.972619      1090.585692   \n",
       "1295   mp-999549       13.421260  13.456055      1038.585697   \n",
       "\n",
       "                                              structure  formula   composition  \n",
       "0     [[2.8943817  2.04663693 5.01321616] Te, [0. 0....     BaTe      (Ba, Te)  \n",
       "1     [[0. 0. 0.] Hf, [-3.78195772 -3.78195772 -3.78...      HfC       (Hf, C)  \n",
       "2     [[0. 0. 0.] Ge, [ 3.45311592  3.45311592 -3.45...      GeC       (Ge, C)  \n",
       "3     [[0.98372595 0.69559929 1.70386332] B, [0. 0. ...      BAs       (B, As)  \n",
       "4               [[0. 0. 0.] Ca, [ 4.95  4.95 -4.95] Se]     CaSe      (Ca, Se)  \n",
       "...                                                 ...      ...           ...  \n",
       "1291  [[1.69099645e-03 3.81913207e+00 1.07685858e-01...  RbPbBr3  (Rb, Pb, Br)  \n",
       "1292  [[1.64439731e-03 3.64832409e+00 1.03287739e-01...  RbPbCl3  (Rb, Pb, Cl)  \n",
       "1293  [[-3.66731982 -1.91142875  2.96640499] K, [ 3....   KGeBr3   (K, Ge, Br)  \n",
       "1294  [[ 1.57631457 -0.32583322 -1.57631457] N, [ 1....       N2           (N)  \n",
       "1295  [[1.46542555 0.84617794 1.95957081] W, [0.0000...      WN2        (W, N)  \n",
       "\n",
       "[1296 rows x 7 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from matminer.featurizers.conversions import StrToComposition\n",
    "\n",
    "stc = StrToComposition()\n",
    "stc.featurize_dataframe(df, \"formula\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed4fb582",
   "metadata": {},
   "source": [
    "## ML models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "c0272201",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "de4938227ed74c94a4b27fede2231cbc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "StrToComposition:   0%|          | 0/1181 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9d6c557245f849d29a19a733b75225c7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "DensityFeatures:   0%|          | 0/1181 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = load_dataset(\"elastic_tensor_2015\")\n",
    "df = df[[\"structure\", \"formula\", \"K_VRH\"]]\n",
    "\n",
    "stc = StrToComposition()\n",
    "df = stc.featurize_dataframe(df, \"formula\")\n",
    "\n",
    "# ep = ElementFraction()\n",
    "# df = ep.featurize_dataframe(df, \"composition\")\n",
    "\n",
    "de = DensityFeatures()\n",
    "df = de.featurize_dataframe(df, \"structure\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "cab7e51b",
   "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>structure</th>\n",
       "      <th>formula</th>\n",
       "      <th>K_VRH</th>\n",
       "      <th>composition</th>\n",
       "      <th>density</th>\n",
       "      <th>vpa</th>\n",
       "      <th>packing fraction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[[0.94814328 2.07280467 2.5112    ] Nb, [5.273...</td>\n",
       "      <td>Nb4CoSi</td>\n",
       "      <td>194.268884</td>\n",
       "      <td>(Nb, Co, Si)</td>\n",
       "      <td>7.834556</td>\n",
       "      <td>16.201654</td>\n",
       "      <td>0.688834</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...</td>\n",
       "      <td>Al(CoSi)2</td>\n",
       "      <td>175.449907</td>\n",
       "      <td>(Al, Co, Si)</td>\n",
       "      <td>5.384968</td>\n",
       "      <td>12.397466</td>\n",
       "      <td>0.644386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]</td>\n",
       "      <td>SiOs</td>\n",
       "      <td>295.077545</td>\n",
       "      <td>(Si, Os)</td>\n",
       "      <td>13.968635</td>\n",
       "      <td>12.976265</td>\n",
       "      <td>0.569426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[[0.         1.09045794 0.84078375] Ga, [0.   ...</td>\n",
       "      <td>Ga</td>\n",
       "      <td>49.130670</td>\n",
       "      <td>(Ga)</td>\n",
       "      <td>6.036267</td>\n",
       "      <td>19.180359</td>\n",
       "      <td>0.479802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[[1.0094265  4.24771709 2.9955487 ] Si, [3.028...</td>\n",
       "      <td>SiRu2</td>\n",
       "      <td>256.768081</td>\n",
       "      <td>(Si, Ru)</td>\n",
       "      <td>9.539514</td>\n",
       "      <td>13.358418</td>\n",
       "      <td>0.598395</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           structure    formula       K_VRH  \\\n",
       "0  [[0.94814328 2.07280467 2.5112    ] Nb, [5.273...    Nb4CoSi  194.268884   \n",
       "1  [[0. 0. 0.] Al, [1.96639263 1.13529553 0.75278...  Al(CoSi)2  175.449907   \n",
       "2   [[1.480346 1.480346 1.480346] Si, [0. 0. 0.] Os]       SiOs  295.077545   \n",
       "3  [[0.         1.09045794 0.84078375] Ga, [0.   ...         Ga   49.130670   \n",
       "4  [[1.0094265  4.24771709 2.9955487 ] Si, [3.028...      SiRu2  256.768081   \n",
       "\n",
       "    composition    density        vpa  packing fraction  \n",
       "0  (Nb, Co, Si)   7.834556  16.201654          0.688834  \n",
       "1  (Al, Co, Si)   5.384968  12.397466          0.644386  \n",
       "2      (Si, Os)  13.968635  12.976265          0.569426  \n",
       "3          (Ga)   6.036267  19.180359          0.479802  \n",
       "4      (Si, Ru)   9.539514  13.358418          0.598395  "
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "6f27fd3f",
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df[\"K_VRH\"]\n",
    "X = df.drop([\"structure\", \"formula\", \"composition\", \"K_VRH\"], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "4a74cfd4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['density', 'vpa', 'packing fraction'], dtype='object')"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "535b0ae5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training RMSE = 12.386 GPa\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "import numpy as np\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "rf = RandomForestRegressor(n_estimators=100, random_state=1)\n",
    "rf.fit(X, y)\n",
    "\n",
    "y_pred = rf.predict(X)\n",
    "mse = mean_squared_error(y, y_pred)\n",
    "print(\"Training RMSE = {:.3f} GPa\".format(np.sqrt(mse)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ad704569",
   "metadata": {},
   "source": [
    "## Cross validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "9aec7bee",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean RMSE: 32.483\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import KFold\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "kfold = KFold(n_splits=10, random_state=1, shuffle=True)\n",
    "scores = cross_val_score(rf, X, y, scoring=\"neg_mean_squared_error\", cv=kfold)\n",
    "\n",
    "rmse_score = [np.sqrt(abs(s)) for s in scores]\n",
    "# rmse_score\n",
    "\n",
    "print(\"Mean RMSE: {:.3f}\".format(np.mean(rmse_score)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "77599ba8",
   "metadata": {},
   "source": [
    "## Model performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "4268c427",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_predict\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "y_pred = cross_val_predict(rf, X, y, cv=kfold)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "3064e6c3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 265,
       "width": 398
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(y, y_pred, 'ro')\n",
    "plt.plot([0, 500], [0, 500], 'k--')\n",
    "plt.xlabel(\"DFT (MP) bulk modulas (GPa)\")\n",
    "plt.ylabel(\"Predicted bulk modulas (GPa)\")\n",
    "plt.xlim(0, 450)\n",
    "plt.ylim(0, 400)\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cec89342",
   "metadata": {},
   "source": [
    "## Model interpretation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "09ac750a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.31108464, 0.60610481, 0.08281055])"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.feature_importances_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "94c14ff4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 437
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "importances = rf.feature_importances_\n",
    "include = X.columns.values\n",
    "indices = np.argsort(importances)[::-1]\n",
    "plt.bar(include[indices], include[indices]) \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2f739111",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}