{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Source of Dataset - https://archive.ics.uci.edu/ml/datasets/Airfoil+Self-Noise#\n",
    "\n",
    "#About Data\n",
    "# NASA data set obtained from a series of aerodynamic and acoustic tests of two and three-dimensional airfoil \n",
    "# blade sections conducted in an anechoic wind tunnel.\n",
    "# The data set comprises different size NACA 0012 airfoils at various wind tunnel speeds and angles of attack. \n",
    "# the span of the airfoil and the observer position were the same in all of the experiments\n",
    "\n",
    "#Objective\n",
    "# Predict Scaled Sound Pressure Level given the attributes correspoding to the experiment."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Load Modules\n",
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Set Directory\n",
    "os.chdir(\"path_to_directory\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Frequency</th>\n",
       "      <th>Angle_of_Attack</th>\n",
       "      <th>Chord_Length</th>\n",
       "      <th>Free_Stream_Velocity</th>\n",
       "      <th>Suction_Side_Displacement</th>\n",
       "      <th>Scaled_Sound_Pressure_Level</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>800</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3048</td>\n",
       "      <td>71.3</td>\n",
       "      <td>0.002663</td>\n",
       "      <td>126.201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3048</td>\n",
       "      <td>71.3</td>\n",
       "      <td>0.002663</td>\n",
       "      <td>125.201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1250</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3048</td>\n",
       "      <td>71.3</td>\n",
       "      <td>0.002663</td>\n",
       "      <td>125.951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1600</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3048</td>\n",
       "      <td>71.3</td>\n",
       "      <td>0.002663</td>\n",
       "      <td>127.591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3048</td>\n",
       "      <td>71.3</td>\n",
       "      <td>0.002663</td>\n",
       "      <td>127.461</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Frequency  Angle_of_Attack  Chord_Length  Free_Stream_Velocity  \\\n",
       "0        800              0.0        0.3048                  71.3   \n",
       "1       1000              0.0        0.3048                  71.3   \n",
       "2       1250              0.0        0.3048                  71.3   \n",
       "3       1600              0.0        0.3048                  71.3   \n",
       "4       2000              0.0        0.3048                  71.3   \n",
       "\n",
       "   Suction_Side_Displacement  Scaled_Sound_Pressure_Level  \n",
       "0                   0.002663                      126.201  \n",
       "1                   0.002663                      125.201  \n",
       "2                   0.002663                      125.951  \n",
       "3                   0.002663                      127.591  \n",
       "4                   0.002663                      127.461  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Load data\n",
    "data = pd.read_csv(\"Data.csv\")\n",
    "data.head() #view snapshot of data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1503 entries, 0 to 1502\n",
      "Data columns (total 6 columns):\n",
      "Frequency                      1503 non-null int64\n",
      "Angle_of_Attack                1503 non-null float64\n",
      "Chord_Length                   1503 non-null float64\n",
      "Free_Stream_Velocity           1503 non-null float64\n",
      "Suction_Side_Displacement      1503 non-null float64\n",
      "Scaled_Sound_Pressure_Level    1503 non-null float64\n",
      "dtypes: float64(5), int64(1)\n",
      "memory usage: 70.5 KB\n"
     ]
    }
   ],
   "source": [
    "#Basic information about data\n",
    "data.info()\n",
    "# There are 1503 observations \n",
    "# no missing values present in the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": false
   },
   "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>Frequency</th>\n",
       "      <th>Angle_of_Attack</th>\n",
       "      <th>Chord_Length</th>\n",
       "      <th>Free_Stream_Velocity</th>\n",
       "      <th>Suction_Side_Displacement</th>\n",
       "      <th>Scaled_Sound_Pressure_Level</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1503.000000</td>\n",
       "      <td>1503.000000</td>\n",
       "      <td>1503.000000</td>\n",
       "      <td>1503.000000</td>\n",
       "      <td>1503.000000</td>\n",
       "      <td>1503.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2886.380572</td>\n",
       "      <td>6.782302</td>\n",
       "      <td>0.136548</td>\n",
       "      <td>50.860745</td>\n",
       "      <td>0.011140</td>\n",
       "      <td>124.835943</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>3152.573137</td>\n",
       "      <td>5.918128</td>\n",
       "      <td>0.093541</td>\n",
       "      <td>15.572784</td>\n",
       "      <td>0.013150</td>\n",
       "      <td>6.898657</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>200.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.025400</td>\n",
       "      <td>31.700000</td>\n",
       "      <td>0.000401</td>\n",
       "      <td>103.380000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>800.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.050800</td>\n",
       "      <td>39.600000</td>\n",
       "      <td>0.002535</td>\n",
       "      <td>120.191000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1600.000000</td>\n",
       "      <td>5.400000</td>\n",
       "      <td>0.101600</td>\n",
       "      <td>39.600000</td>\n",
       "      <td>0.004957</td>\n",
       "      <td>125.721000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4000.000000</td>\n",
       "      <td>9.900000</td>\n",
       "      <td>0.228600</td>\n",
       "      <td>71.300000</td>\n",
       "      <td>0.015576</td>\n",
       "      <td>129.995500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>20000.000000</td>\n",
       "      <td>22.200000</td>\n",
       "      <td>0.304800</td>\n",
       "      <td>71.300000</td>\n",
       "      <td>0.058411</td>\n",
       "      <td>140.987000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Frequency  Angle_of_Attack  Chord_Length  Free_Stream_Velocity  \\\n",
       "count   1503.000000      1503.000000   1503.000000           1503.000000   \n",
       "mean    2886.380572         6.782302      0.136548             50.860745   \n",
       "std     3152.573137         5.918128      0.093541             15.572784   \n",
       "min      200.000000         0.000000      0.025400             31.700000   \n",
       "25%      800.000000         2.000000      0.050800             39.600000   \n",
       "50%     1600.000000         5.400000      0.101600             39.600000   \n",
       "75%     4000.000000         9.900000      0.228600             71.300000   \n",
       "max    20000.000000        22.200000      0.304800             71.300000   \n",
       "\n",
       "       Suction_Side_Displacement  Scaled_Sound_Pressure_Level  \n",
       "count                1503.000000                  1503.000000  \n",
       "mean                    0.011140                   124.835943  \n",
       "std                     0.013150                     6.898657  \n",
       "min                     0.000401                   103.380000  \n",
       "25%                     0.002535                   120.191000  \n",
       "50%                     0.004957                   125.721000  \n",
       "75%                     0.015576                   129.995500  \n",
       "max                     0.058411                   140.987000  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Exploratory Analysis\n",
    "##Univariate Analysis\n",
    "\n",
    "#get data summary\n",
    "data.describe()\n",
    "# This informs us about the distribution of each variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x1440 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Histogram\n",
    "data.hist(figsize=(16, 20), bins=10, xlabelsize=8, ylabelsize=8);\n",
    "\n",
    "# most of the variables seem right skewed but scaled_sound_pressure_level has left skewed normal distribution\n",
    "# Also, some variables like chord_length attain values in certain intervals\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "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>Frequency</th>\n",
       "      <th>Angle_of_Attack</th>\n",
       "      <th>Chord_Length</th>\n",
       "      <th>Free_Stream_Velocity</th>\n",
       "      <th>Suction_Side_Displacement</th>\n",
       "      <th>Scaled_Sound_Pressure_Level</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Frequency</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.272765</td>\n",
       "      <td>-0.003661</td>\n",
       "      <td>0.133664</td>\n",
       "      <td>-0.230107</td>\n",
       "      <td>-0.390711</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Angle_of_Attack</th>\n",
       "      <td>-0.272765</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.504868</td>\n",
       "      <td>0.058760</td>\n",
       "      <td>0.753394</td>\n",
       "      <td>-0.156108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Chord_Length</th>\n",
       "      <td>-0.003661</td>\n",
       "      <td>-0.504868</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.003787</td>\n",
       "      <td>-0.220842</td>\n",
       "      <td>-0.236162</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Free_Stream_Velocity</th>\n",
       "      <td>0.133664</td>\n",
       "      <td>0.058760</td>\n",
       "      <td>0.003787</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.003974</td>\n",
       "      <td>0.125103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Suction_Side_Displacement</th>\n",
       "      <td>-0.230107</td>\n",
       "      <td>0.753394</td>\n",
       "      <td>-0.220842</td>\n",
       "      <td>-0.003974</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.312670</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Scaled_Sound_Pressure_Level</th>\n",
       "      <td>-0.390711</td>\n",
       "      <td>-0.156108</td>\n",
       "      <td>-0.236162</td>\n",
       "      <td>0.125103</td>\n",
       "      <td>-0.312670</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             Frequency  Angle_of_Attack  Chord_Length  \\\n",
       "Frequency                     1.000000        -0.272765     -0.003661   \n",
       "Angle_of_Attack              -0.272765         1.000000     -0.504868   \n",
       "Chord_Length                 -0.003661        -0.504868      1.000000   \n",
       "Free_Stream_Velocity          0.133664         0.058760      0.003787   \n",
       "Suction_Side_Displacement    -0.230107         0.753394     -0.220842   \n",
       "Scaled_Sound_Pressure_Level  -0.390711        -0.156108     -0.236162   \n",
       "\n",
       "                             Free_Stream_Velocity  Suction_Side_Displacement  \\\n",
       "Frequency                                0.133664                  -0.230107   \n",
       "Angle_of_Attack                          0.058760                   0.753394   \n",
       "Chord_Length                             0.003787                  -0.220842   \n",
       "Free_Stream_Velocity                     1.000000                  -0.003974   \n",
       "Suction_Side_Displacement               -0.003974                   1.000000   \n",
       "Scaled_Sound_Pressure_Level              0.125103                  -0.312670   \n",
       "\n",
       "                             Scaled_Sound_Pressure_Level  \n",
       "Frequency                                      -0.390711  \n",
       "Angle_of_Attack                                -0.156108  \n",
       "Chord_Length                                   -0.236162  \n",
       "Free_Stream_Velocity                            0.125103  \n",
       "Suction_Side_Displacement                      -0.312670  \n",
       "Scaled_Sound_Pressure_Level                     1.000000  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "##Bivariate Analysis\n",
    "corr_matrix = data.corr()\n",
    "corr_matrix\n",
    "\n",
    "# As suspected from the distribution, we have negative correlation for most of the variables with the dependent variable\n",
    "# also, there is good correlation between Angle_of_Attack and Suction_Side_Displacement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x216 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Scatter Plots\n",
    "\n",
    "fig, axes = plt.subplots(1, 5, figsize=(15,3))\n",
    "\n",
    "for i in range(5):\n",
    "    axes[i].scatter(data.iloc[:,i], data[\"Scaled_Sound_Pressure_Level\"])\n",
    "    axes[i].set_title(data.columns[i])\n",
    "    \n",
    "# There does not seem to be a clear linear relationship present \n",
    "# There is no obvious transformation from the plots that would give us linear relationship"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#split data into dependent and independent variables\n",
    "X = data.drop(['Scaled_Sound_Pressure_Level'],axis=  1)\n",
    "Y = data.loc[:, \"Scaled_Sound_Pressure_Level\"].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#split data into train and test\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size = 0.2, random_state = 0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fitting Multiple Linear Regression to the Training set\n",
    "from sklearn.linear_model import LinearRegression\n",
    "regressor = LinearRegression()\n",
    "regressor.fit(X_train, y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Predicting the Test set results\n",
    "y_pred = regressor.predict(X_test)\n",
    "\n",
    "#Getting Fitted values from model\n",
    "y_fitted = regressor.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Coefficients: [-1.28041334e-03 -4.10769342e-01 -3.60326340e+01  9.81537999e-02\n",
      " -1.44109711e+02]\n",
      "Intercept: 132.8456572124818\n",
      "Root mean squared error:  23.618448776410744\n",
      "R2 score:  0.5047490146008985\n",
      "Adj R2 score:  0.5026785673375243\n"
     ]
    }
   ],
   "source": [
    "#model evaluation train\n",
    "from sklearn.metrics import mean_squared_error, r2_score\n",
    "\n",
    "rmse_train = mean_squared_error(y_train, y_fitted)\n",
    "r2_train = r2_score(y_train, y_fitted)\n",
    "adj_r2_train = 1 - float(len(y_train)-1)/(len(y_train)-len(regressor.coef_)-1)*(1 - r2_train)\n",
    "\n",
    "print('Coefficients:' ,regressor.coef_)\n",
    "print('Intercept:', regressor.intercept_)\n",
    "print('Root mean squared error: ', rmse_train)\n",
    "print('R2 score: ', r2_train)\n",
    "print('Adj R2 score: ', adj_r2_train)\n",
    "\n",
    "# R2 value is pretty low owing to not so strong linear relationshipspresent in the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Root mean squared error:  20.765101495623412\n"
     ]
    }
   ],
   "source": [
    "#model evaluation test\n",
    "from sklearn.metrics import mean_squared_error, r2_score\n",
    "\n",
    "rmse_test = mean_squared_error(y_test, y_pred)\n",
    "\n",
    "print('Root mean squared error: ', rmse_test)\n",
    "\n",
    "# Test RMSE is slightly lesser than train RMSE suggesting model is not overfitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Assumptions Testing\n",
    "\n",
    "#1 - Linearity - We already checked for linear relationship in exploratory analsyis above\n",
    "#2 - Normality of Residuals\n",
    "\n",
    "residuals = y_train - y_fitted\n",
    "\n",
    "plt.hist(residuals,bins = 10)\n",
    "plt.show()\n",
    "# Residuals are following Normal Distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#3 - Homoscedasticity (Constant Variance)\n",
    "\n",
    "plt.scatter(y_fitted,residuals)\n",
    "plt.show()\n",
    "# There does not seem to be any pattern in the plot, also, it seems like a constant variance plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#4 - Independence of Residuals\n",
    "\n",
    "plt.scatter(range(residuals.shape[0]),residuals)\n",
    "plt.show()\n",
    "# The plot is completely random and no pattern can be found"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Hence, all the assumptions are passed. \n",
    "# There is lot of scope of improvements in the model with some variable transformation \n",
    "# and if any domain knowledge can be applied."
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}