Pangilinan_Final_Project.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pymatgen as pym\n",
    "from pymatgen import MPRester\n",
    "\n",
    "\"\"\"The above will not be used in this particular notebook per se, but it was included as it was used to \n",
    "generate band gap data from the Materials API, and was also used to calculate the features.\n",
    "\"\"\"\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Classic importing section\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn import metrics\n",
    "\n",
    "from sklearn import linear_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split, cross_validate, cross_val_predict\n",
    "from sklearn.model_selection import KFold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "from sklearn.model_selection import train_test_split, cross_validate, cross_val_predict\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.linear_model import Ridge, RidgeCV\n",
    "\n",
    "from sklearn import metrics\n",
    "from sklearn.model_selection import KFold, GridSearchCV\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor, ExtraTreesRegressor\n",
    "\n",
    "from sklearn "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Importing the data.\n",
    "\n",
    "\n",
    "data = pd.read_csv('binary_compounds.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>Unnamed: 0</th>\n",
       "      <th>pretty_formula</th>\n",
       "      <th>band_gap</th>\n",
       "      <th>comp</th>\n",
       "      <th>eneg_diff</th>\n",
       "      <th>atom_frac</th>\n",
       "      <th>molec_weight</th>\n",
       "      <th>group_num</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>BN</td>\n",
       "      <td>5.6697</td>\n",
       "      <td>B1 N1</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.50</td>\n",
       "      <td>24.817700</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>TlN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Tl1 N1</td>\n",
       "      <td>1.42</td>\n",
       "      <td>0.50</td>\n",
       "      <td>218.390000</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>GaN</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Ga1 N1</td>\n",
       "      <td>1.23</td>\n",
       "      <td>0.50</td>\n",
       "      <td>83.729700</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>AlN</td>\n",
       "      <td>3.4683</td>\n",
       "      <td>Al1 N1</td>\n",
       "      <td>1.43</td>\n",
       "      <td>0.50</td>\n",
       "      <td>40.988239</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>18</td>\n",
       "      <td>InN</td>\n",
       "      <td>0.4777</td>\n",
       "      <td>In1 N1</td>\n",
       "      <td>1.26</td>\n",
       "      <td>0.50</td>\n",
       "      <td>128.824700</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>713</th>\n",
       "      <td>713</td>\n",
       "      <td>Sn3Te</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Sn3 Te1</td>\n",
       "      <td>3.78</td>\n",
       "      <td>0.75</td>\n",
       "      <td>483.730000</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>714</th>\n",
       "      <td>714</td>\n",
       "      <td>TePb3</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Te1 Pb3</td>\n",
       "      <td>4.89</td>\n",
       "      <td>0.25</td>\n",
       "      <td>749.200000</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>717</th>\n",
       "      <td>717</td>\n",
       "      <td>SnTe3</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Sn1 Te3</td>\n",
       "      <td>4.34</td>\n",
       "      <td>0.25</td>\n",
       "      <td>501.510000</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>719</th>\n",
       "      <td>719</td>\n",
       "      <td>SnTe</td>\n",
       "      <td>0.6624</td>\n",
       "      <td>Sn1 Te1</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.50</td>\n",
       "      <td>246.310000</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>720</th>\n",
       "      <td>720</td>\n",
       "      <td>Ge3Te</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>Ge3 Te1</td>\n",
       "      <td>3.93</td>\n",
       "      <td>0.75</td>\n",
       "      <td>345.520000</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>183 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Unnamed: 0 pretty_formula  band_gap     comp  eneg_diff  atom_frac  \\\n",
       "0             0             BN    5.6697    B1 N1       1.00       0.50   \n",
       "1             1            TlN    0.0000   Tl1 N1       1.42       0.50   \n",
       "3             3            GaN    0.0000   Ga1 N1       1.23       0.50   \n",
       "7             7            AlN    3.4683   Al1 N1       1.43       0.50   \n",
       "18           18            InN    0.4777   In1 N1       1.26       0.50   \n",
       "..          ...            ...       ...      ...        ...        ...   \n",
       "713         713          Sn3Te    0.0000  Sn3 Te1       3.78       0.75   \n",
       "714         714          TePb3    0.0000  Te1 Pb3       4.89       0.25   \n",
       "717         717          SnTe3    0.0000  Sn1 Te3       4.34       0.25   \n",
       "719         719           SnTe    0.6624  Sn1 Te1       0.14       0.50   \n",
       "720         720          Ge3Te    0.0000  Ge3 Te1       3.93       0.75   \n",
       "\n",
       "     molec_weight  group_num  \n",
       "0       24.817700         13  \n",
       "1      218.390000         13  \n",
       "3       83.729700         13  \n",
       "7       40.988239         13  \n",
       "18     128.824700         13  \n",
       "..            ...        ...  \n",
       "713    483.730000         14  \n",
       "714    749.200000         16  \n",
       "717    501.510000         14  \n",
       "719    246.310000         14  \n",
       "720    345.520000         14  \n",
       "\n",
       "[183 rows x 8 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Some data processing:\n",
    "\n",
    "\n",
    "df_wcut = data[data['molec_weight']<1000]   #Removing any very heavy chemical compounds\n",
    "\n",
    "\n",
    "df_ecut = df_wcut[df_wcut['eneg_diff']<10]  \n",
    "\n",
    "#Considering only molecules with relatively small electronegativity differences\n",
    "\n",
    "\n",
    "df = df_ecut.drop_duplicates(subset=['pretty_formula'])  \n",
    "\n",
    "\"\"\"Removing any duplicate molecules. Originally tried, using all 700+, duplicates included... \n",
    "but that gave very low scores for both\n",
    "\"\"\"\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>eneg_diff</th>\n",
       "      <th>atom_frac</th>\n",
       "      <th>molec_weight</th>\n",
       "      <th>group_num</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.00</td>\n",
       "      <td>0.50</td>\n",
       "      <td>24.817700</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.42</td>\n",
       "      <td>0.50</td>\n",
       "      <td>218.390000</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.23</td>\n",
       "      <td>0.50</td>\n",
       "      <td>83.729700</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1.43</td>\n",
       "      <td>0.50</td>\n",
       "      <td>40.988239</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>1.26</td>\n",
       "      <td>0.50</td>\n",
       "      <td>128.824700</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>713</th>\n",
       "      <td>3.78</td>\n",
       "      <td>0.75</td>\n",
       "      <td>483.730000</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>714</th>\n",
       "      <td>4.89</td>\n",
       "      <td>0.25</td>\n",
       "      <td>749.200000</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>717</th>\n",
       "      <td>4.34</td>\n",
       "      <td>0.25</td>\n",
       "      <td>501.510000</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>719</th>\n",
       "      <td>0.14</td>\n",
       "      <td>0.50</td>\n",
       "      <td>246.310000</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>720</th>\n",
       "      <td>3.93</td>\n",
       "      <td>0.75</td>\n",
       "      <td>345.520000</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>183 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     eneg_diff  atom_frac  molec_weight  group_num\n",
       "0         1.00       0.50     24.817700         13\n",
       "1         1.42       0.50    218.390000         13\n",
       "3         1.23       0.50     83.729700         13\n",
       "7         1.43       0.50     40.988239         13\n",
       "18        1.26       0.50    128.824700         13\n",
       "..         ...        ...           ...        ...\n",
       "713       3.78       0.75    483.730000         14\n",
       "714       4.89       0.25    749.200000         16\n",
       "717       4.34       0.25    501.510000         14\n",
       "719       0.14       0.50    246.310000         14\n",
       "720       3.93       0.75    345.520000         14\n",
       "\n",
       "[183 rows x 4 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features = df[['eneg_diff','atom_frac','molec_weight','group_num']] #defining the features data frame\n",
    " \n",
    "features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      5.6697\n",
       "1      0.0000\n",
       "3      0.0000\n",
       "7      3.4683\n",
       "18     0.4777\n",
       "        ...  \n",
       "713    0.0000\n",
       "714    0.0000\n",
       "717    0.0000\n",
       "719    0.6624\n",
       "720    0.0000\n",
       "Name: band_gap, Length: 183, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target = df['band_gap']  #defining the band gap data as the target\n",
    "\n",
    "target"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### First, let's visualize the data.\n",
    "\n",
    "Here I make 3 graphs, to see if there is any noticeable trend when comparing band gaps to the Electronegativity Difference, the Molecular Weight, and the Atomic Fraction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      5.6697\n",
       "1      0.0000\n",
       "3      0.0000\n",
       "7      3.4683\n",
       "18     0.4777\n",
       "        ...  \n",
       "713    0.0000\n",
       "714    0.0000\n",
       "717    0.0000\n",
       "719    0.6624\n",
       "720    0.0000\n",
       "Name: band_gap, Length: 183, dtype: float64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y = target\n",
    "\n",
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "\n",
    "x1 = features['eneg_diff'].to_numpy()\n",
    "\n",
    "x2 = features['atom_frac'].to_numpy()\n",
    "\n",
    "x3 = features['molec_weight'].to_numpy()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Band Gap')"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#A graph examining the relationship between the electronegativty difference and the band gap\n",
    "\n",
    "\n",
    "figure_one = plt.scatter(x1, y) \n",
    "plt.xlabel('Electronegativity Difference')\n",
    "plt.ylabel('Band Gap') \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Band Gap')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#A graph examining the relationship between the atomic fraction of the cation and the band gap\n",
    "\n",
    "\n",
    "figure_two = plt.scatter(x2, y) \n",
    "plt.xlabel('Atomic Fraction')\n",
    "plt.ylabel('Band Gap') \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#A graph examining the relationship between the molecular weight of the compound and the band gap\n",
    "\n",
    "\n",
    "figure_three = plt.scatter(x3, y) \n",
    "plt.xlabel('Molecular Weight')\n",
    "plt.ylabel('Band Gap')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### First, we will try to see how a RidgeRegression model performs\n",
    "\n",
    "While playing around, I also used the Linear Regression without Regularization, but it certainly didn't do too strongly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = RidgeCV()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "cv = KFold(n_splits=5, shuffle = True, random_state=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'fit_time': array([0.00399041, 0.00498581, 0.00398993, 0.00395679, 0.00298619]),\n",
       " 'score_time': array([0.00199461, 0.00199413, 0.00099659, 0.00199556, 0.00099945]),\n",
       " 'test_score': array([0.36974692, 0.29593865, 0.53236762, 0.22267275, 0.34149903]),\n",
       " 'train_score': array([0.41692426, 0.41719967, 0.3791842 , 0.44244848, 0.40454511])}"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores = cross_validate(model, features, target, cv = cv, return_train_score = True)\n",
    "\n",
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.352 0.103\n",
      "0.412 0.021\n"
     ]
    }
   ],
   "source": [
    "print('{:.3f}'.format(scores['test_score'].mean()), '{:.3f}'.format(scores['test_score'].std()))\n",
    "print('{:.3f}'.format(scores['train_score'].mean()), '{:.3f}'.format(scores['train_score'].std()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Clearly, a very high bias issue. To see whether or not adding more data would be beneficial, it is useful to plot a learning curve."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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>test_score</th>\n",
       "      <th>train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.253761</td>\n",
       "      <td>0.178318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.336597</td>\n",
       "      <td>0.368044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.328913</td>\n",
       "      <td>0.351448</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.297636</td>\n",
       "      <td>0.389622</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.352445</td>\n",
       "      <td>0.412060</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   test_score  train_score\n",
       "0   -0.253761     0.178318\n",
       "1    0.336597     0.368044\n",
       "2    0.328913     0.351448\n",
       "3    0.297636     0.389622\n",
       "4    0.352445     0.412060"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''Obtaining data, for the learning curve, by examining the model beginning with 1/5th of the data \n",
    "and increasing the size by another 1/5th.\n",
    "'''\n",
    "\n",
    "size = np.shape(features)\n",
    "\n",
    "test_data = pd.DataFrame(columns = ('test_score','train_score'))\n",
    "\n",
    "for i in range(1,6):\n",
    "    per_feat = round(i/5*size[0])\n",
    "\n",
    "    scores = cross_validate(model, features.iloc[:per_feat,:], target[:per_feat], cv = cv, return_train_score = True)\n",
    "    \n",
    "    df2 = pd.DataFrame([[scores['test_score'].mean(), scores['train_score'].mean()]],columns = ('test_score','train_score'))\n",
    "        \n",
    "    test_data = test_data.append(df2, ignore_index = True)\n",
    "    \n",
    "\n",
    "test_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0   -0.253761\n",
       "1    0.336597\n",
       "2    0.328913\n",
       "3    0.297636\n",
       "4    0.352445\n",
       "Name: test_score, dtype: float64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_scores = test_data['test_score']\n",
    "\n",
    "test_scores\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.178318\n",
       "1    0.368044\n",
       "2    0.351448\n",
       "3    0.389622\n",
       "4    0.412060\n",
       "Name: train_score, dtype: float64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_scores = test_data['train_score']\n",
    "\n",
    "train_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Learning Curve')"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plotting the learning curve\n",
    "\n",
    "\n",
    "x = np.array([round(1/5*size[0]), round(2/5*size[0]), round(3/5*size[0]), round(4/5*size[0]), size[0]])\n",
    "\n",
    "y1 = test_scores\n",
    "y2 = train_scores\n",
    "\n",
    "plt.plot(x, y1, label = 'Test Scores')\n",
    "plt.plot(x, y2, label = 'Training Score')\n",
    "plt.xlabel('Training Size')\n",
    "plt.ylabel('Score')\n",
    "plt.title('Learning Curve')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler , RobustScaler"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As can see from the graph above, it doesn't appear that adding data may necessarily help. However, to match the paper, which used 4096 compounds, perhaps more would give a better picture.\n",
    "\n",
    "In spite of this, in the paper, they also happened to scale some of the features."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Now let's try a more powerful method, the Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = RandomForestRegressor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'fit_time': array([0.12144947, 0.10971832, 0.11264896, 0.10776067, 0.09773684]),\n",
       " 'score_time': array([0.00698113, 0.00598311, 0.00702429, 0.00597262, 0.00599241]),\n",
       " 'test_score': array([0.69552454, 0.58445841, 0.8874252 , 0.70898738, 0.72534871]),\n",
       " 'train_score': array([0.96463456, 0.95983126, 0.96236467, 0.96594984, 0.96056479])}"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores = cross_validate(model, features, target , cv = cv, return_train_score=True)\n",
    "\n",
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.720348846226546, 0.9626690218610457)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(scores['test_score']), np.mean(scores['train_score'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The model seems to perform much better as compared to the Linear Model. However, there is now a variance issue that must be taken care of. Like before, let's see how the learning curve looks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "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>test_score</th>\n",
       "      <th>train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.485289</td>\n",
       "      <td>0.905163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.470413</td>\n",
       "      <td>0.940884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.231276</td>\n",
       "      <td>0.935591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.669536</td>\n",
       "      <td>0.958989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.716763</td>\n",
       "      <td>0.962224</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   test_score  train_score\n",
       "0    0.485289     0.905163\n",
       "1    0.470413     0.940884\n",
       "2    0.231276     0.935591\n",
       "3    0.669536     0.958989\n",
       "4    0.716763     0.962224"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#same as prior\n",
    "\n",
    "test_data = pd.DataFrame(columns = ('test_score','train_score'))\n",
    "\n",
    "for i in range(1,6):\n",
    "    per_feat = round(i/5*size[0])\n",
    "\n",
    "    scores = cross_validate(model, features.iloc[:per_feat,:], target[:per_feat], cv = cv, return_train_score = True)\n",
    "    \n",
    "    df2 = pd.DataFrame([[scores['test_score'].mean(), scores['train_score'].mean()]],columns = ('test_score','train_score'))\n",
    "        \n",
    "    test_data = test_data.append(df2, ignore_index = True)\n",
    "    \n",
    "\n",
    "test_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_scores = test_data['test_score']\n",
    "\n",
    "train_scores = test_data['train_score']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Learning Curve')"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plotting the learning curve\n",
    "\n",
    "\n",
    "x = np.array([round(1/5*size[0]), round(2/5*size[0]), round(3/5*size[0]), round(4/5*size[0]), size[0]])\n",
    "\n",
    "y1 = test_scores\n",
    "y2 = train_scores\n",
    "\n",
    "plt.plot(x, y1, label = 'Test Scores')\n",
    "plt.plot(x, y2, label = 'Training Score')\n",
    "plt.xlabel('Training Set Size')\n",
    "plt.ylabel('Score')\n",
    "plt.title('Learning Curve')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is a strange dip when the size of the training data is about 3/5th of the overall amount of samples. Despite this, there seems to be a general trend of the score improving as more data is added.\n",
    "\n",
    "Now let's examine the parameters themselves to see if that helps."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 162 candidates, totalling 810 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=4)]: Using backend LokyBackend with 4 concurrent workers.\n",
      "[Parallel(n_jobs=4)]: Done  58 tasks      | elapsed:    2.7s\n",
      "[Parallel(n_jobs=4)]: Done 300 tasks      | elapsed:   13.5s\n",
      "[Parallel(n_jobs=4)]: Done 706 tasks      | elapsed:   32.2s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best params, best score: 0.7441 {'max_depth': None, 'max_features': 2, 'min_samples_leaf': 1, 'min_samples_split': 2, 'n_estimators': 100}\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=4)]: Done 810 out of 810 | elapsed:   37.1s finished\n"
     ]
    }
   ],
   "source": [
    "#Performing a Grid Search to see if we can obtain the best model. This takes a couple of minutes.\n",
    "\n",
    "\n",
    "\n",
    "parameters = {'max_depth':[3, 6, None], \\\n",
    "              'max_features':[None,2], 'n_estimators':[50, 100, 200], 'min_samples_leaf': [1,3,5],'min_samples_split': [2,4,6]}\n",
    "nmodels = np.product([len(el) for el in parameters.values()])\n",
    "model = GridSearchCV(RandomForestRegressor(random_state = 10), parameters, cv = cv, \\\n",
    "                     verbose = 2, n_jobs = 4, return_train_score=True)\n",
    "model.fit(features,target)\n",
    "\n",
    "print('Best params, best score:', \"{:.4f}\".format(model.best_score_), \\\n",
    "      model.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "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>params</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>mean_train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>136</th>\n",
       "      <td>{'max_depth': None, 'max_features': 2, 'min_sa...</td>\n",
       "      <td>0.744114</td>\n",
       "      <td>0.080753</td>\n",
       "      <td>0.966130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>{'max_depth': None, 'max_features': 2, 'min_sa...</td>\n",
       "      <td>0.740855</td>\n",
       "      <td>0.079909</td>\n",
       "      <td>0.964897</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>138</th>\n",
       "      <td>{'max_depth': None, 'max_features': 2, 'min_sa...</td>\n",
       "      <td>0.736206</td>\n",
       "      <td>0.096220</td>\n",
       "      <td>0.942114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>139</th>\n",
       "      <td>{'max_depth': None, 'max_features': 2, 'min_sa...</td>\n",
       "      <td>0.733731</td>\n",
       "      <td>0.091449</td>\n",
       "      <td>0.944712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82</th>\n",
       "      <td>{'max_depth': 6, 'max_features': 2, 'min_sampl...</td>\n",
       "      <td>0.731210</td>\n",
       "      <td>0.085323</td>\n",
       "      <td>0.929815</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>{'max_depth': 3, 'max_features': None, 'min_sa...</td>\n",
       "      <td>0.573607</td>\n",
       "      <td>0.144099</td>\n",
       "      <td>0.732311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>{'max_depth': 3, 'max_features': None, 'min_sa...</td>\n",
       "      <td>0.573607</td>\n",
       "      <td>0.144099</td>\n",
       "      <td>0.732311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>{'max_depth': 3, 'max_features': None, 'min_sa...</td>\n",
       "      <td>0.570051</td>\n",
       "      <td>0.139797</td>\n",
       "      <td>0.729352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>{'max_depth': 3, 'max_features': None, 'min_sa...</td>\n",
       "      <td>0.570051</td>\n",
       "      <td>0.139797</td>\n",
       "      <td>0.729352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>{'max_depth': 3, 'max_features': None, 'min_sa...</td>\n",
       "      <td>0.570051</td>\n",
       "      <td>0.139797</td>\n",
       "      <td>0.729352</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>162 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                params  mean_test_score  \\\n",
       "136  {'max_depth': None, 'max_features': 2, 'min_sa...         0.744114   \n",
       "135  {'max_depth': None, 'max_features': 2, 'min_sa...         0.740855   \n",
       "138  {'max_depth': None, 'max_features': 2, 'min_sa...         0.736206   \n",
       "139  {'max_depth': None, 'max_features': 2, 'min_sa...         0.733731   \n",
       "82   {'max_depth': 6, 'max_features': 2, 'min_sampl...         0.731210   \n",
       "..                                                 ...              ...   \n",
       "23   {'max_depth': 3, 'max_features': None, 'min_sa...         0.573607   \n",
       "26   {'max_depth': 3, 'max_features': None, 'min_sa...         0.573607   \n",
       "18   {'max_depth': 3, 'max_features': None, 'min_sa...         0.570051   \n",
       "21   {'max_depth': 3, 'max_features': None, 'min_sa...         0.570051   \n",
       "24   {'max_depth': 3, 'max_features': None, 'min_sa...         0.570051   \n",
       "\n",
       "     std_test_score  mean_train_score  \n",
       "136        0.080753          0.966130  \n",
       "135        0.079909          0.964897  \n",
       "138        0.096220          0.942114  \n",
       "139        0.091449          0.944712  \n",
       "82         0.085323          0.929815  \n",
       "..              ...               ...  \n",
       "23         0.144099          0.732311  \n",
       "26         0.144099          0.732311  \n",
       "18         0.139797          0.729352  \n",
       "21         0.139797          0.729352  \n",
       "24         0.139797          0.729352  \n",
       "\n",
       "[162 rows x 4 columns]"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Checking which set of parameters gave the best model\n",
    "\n",
    "scores = pd.DataFrame(model.cv_results_)\n",
    "scoresCV = scores[['params','mean_test_score','std_test_score','mean_train_score']].sort_values(by = 'mean_test_score', \\\n",
    "                                                    ascending = False)\n",
    "\n",
    "scoresCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Taking the best model from the grid search done above.\n",
    "\n",
    "bm = model.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'fit_time': array([0.11758876, 0.09473562, 0.09375048, 0.09574437, 0.09375   ]),\n",
       " 'score_time': array([0.00598025, 0.00598407, 0.00598288, 0.00598454, 0.00694084]),\n",
       " 'test_score': array([0.66874202, 0.66281451, 0.88605936, 0.75683468, 0.74611814]),\n",
       " 'train_score': array([0.96882373, 0.96836708, 0.96535962, 0.96466457, 0.96343454])}"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores = cross_validate(bm, features, target , cv = cv, return_train_score=True)\n",
    "\n",
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.7441137422147153, 0.9661299078181175)"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(scores['test_score']), np.mean(scores['train_score'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ultimately, there doesn't appear to be too much improvement if any, at least based off of the parameters used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "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>test_score</th>\n",
       "      <th>train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.428967</td>\n",
       "      <td>0.923015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.569455</td>\n",
       "      <td>0.946734</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.322057</td>\n",
       "      <td>0.940518</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.676266</td>\n",
       "      <td>0.963045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.744114</td>\n",
       "      <td>0.966130</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   test_score  train_score\n",
       "0    0.428967     0.923015\n",
       "1    0.569455     0.946734\n",
       "2    0.322057     0.940518\n",
       "3    0.676266     0.963045\n",
       "4    0.744114     0.966130"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#same as prior\n",
    "\n",
    "test_data = pd.DataFrame(columns = ('test_score','train_score'))\n",
    "\n",
    "for i in range(1,6):\n",
    "    per_feat = round(i/5*size[0])\n",
    "\n",
    "    scores = cross_validate(bm, features.iloc[:per_feat,:], target[:per_feat], cv = cv, return_train_score = True)\n",
    "    \n",
    "    df2 = pd.DataFrame([[scores['test_score'].mean(), scores['train_score'].mean()]],columns = ('test_score','train_score'))\n",
    "        \n",
    "    test_data = test_data.append(df2, ignore_index = True)\n",
    "    \n",
    "\n",
    "test_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_scores = test_data['test_score']\n",
    "\n",
    "train_scores = test_data['train_score']\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### References:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. Natarajan, AS, Varadhan R, Ezhilvel ME. 2016. Band Gap Estimation Using Machine Learning Techniques. Vixra.org\n",
    "\n",
    "2. J.M.K.C. Donev et al. (2015). Energy Education - Band gap [Online]. Available: https://energyeducation.ca/encyclopedia/Band_gap. \n",
    "\n",
    "3. Zhang, Y., Ling C. 2018. A strategy to apply machine learning to small data sets in materials sciences. Nature\n",
    "\n",
    "4. Ong, S. P.; Cholia, S.; Jain, A.; Brafman, M.; Gunter, D.; Ceder, G.; Persson, K. a. The Materials Application Programming Interface (API): A simple, flexible and efficient API for materials data based on Representational State Transfer (REST) principles, Comput. Mater. Sci., 2015, 97, 209–215. doi:10.1016/j.commatsci.2014.10.037.\n",
    "\n",
    "5. Some Previous Notebooks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}