Linear Regression_ppp.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#Linear Regression for cancer prediction\n",
    "# coding from scratch without using scikit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "######################################################################################## \n",
      " Performing Linear regression on dataset 1\n",
      " Corelating age specific rate and the case count from a dataset\n",
      "(19, 6)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CancerType</th>\n",
       "      <th>Year</th>\n",
       "      <th>Age</th>\n",
       "      <th>AgeSpecificRate</th>\n",
       "      <th>CaseCount</th>\n",
       "      <th>Population</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>'All Cancer Sites Combined'</td>\n",
       "      <td>'2016'</td>\n",
       "      <td>'1-4'</td>\n",
       "      <td>22.1</td>\n",
       "      <td>3535</td>\n",
       "      <td>'15964200'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>'All Cancer Sites Combined'</td>\n",
       "      <td>'2016'</td>\n",
       "      <td>'10-14'</td>\n",
       "      <td>15.6</td>\n",
       "      <td>3224</td>\n",
       "      <td>'20627202'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>'All Cancer Sites Combined'</td>\n",
       "      <td>'2016'</td>\n",
       "      <td>'15-19'</td>\n",
       "      <td>24.0</td>\n",
       "      <td>5084</td>\n",
       "      <td>'21151384'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>'All Cancer Sites Combined'</td>\n",
       "      <td>'2016'</td>\n",
       "      <td>'20-24'</td>\n",
       "      <td>36.0</td>\n",
       "      <td>8051</td>\n",
       "      <td>'22384013'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>'All Cancer Sites Combined'</td>\n",
       "      <td>'2016'</td>\n",
       "      <td>'25-29'</td>\n",
       "      <td>60.2</td>\n",
       "      <td>13797</td>\n",
       "      <td>'22931428'</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    CancerType    Year      Age  AgeSpecificRate  CaseCount  \\\n",
       "0  'All Cancer Sites Combined'  '2016'    '1-4'             22.1       3535   \n",
       "1  'All Cancer Sites Combined'  '2016'  '10-14'             15.6       3224   \n",
       "2  'All Cancer Sites Combined'  '2016'  '15-19'             24.0       5084   \n",
       "3  'All Cancer Sites Combined'  '2016'  '20-24'             36.0       8051   \n",
       "4  'All Cancer Sites Combined'  '2016'  '25-29'             60.2      13797   \n",
       "\n",
       "   Population  \n",
       "0  '15964200'  \n",
       "1  '20627202'  \n",
       "2  '21151384'  \n",
       "3  '22384013'  \n",
       "4  '22931428'  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#reading data from csv\n",
    "#importing libraries\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "plt.rcParams['figure.figsize']=(20.0,10.0)\n",
    "data = pd.read_csv('USCS_DemographicAge(lin).csv')\n",
    "#to have a heads up about the data we are dealing with\n",
    "\n",
    "print (\"######################################################################################## \\n Performing Linear regression on dataset 1\\n Corelating age specific rate and the case count from a dataset\")\n",
    "print(data.shape)\n",
    "#look at the first five rows of data for easier programming\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[   22.1    15.6    24.     36.     60.2    98.    146.1   227.4   348.5\n",
      "    13.3   560.9   829.3  1156.3  1571.8  1893.7  2117.3  2198.3  1982.4\n",
      "    28.1]\n",
      "[  3535   3224   5084   8051  13797  21412  30418  44872  73074   2726\n",
      " 122556 182339 225365 264524 223764 177317 129034 126512   1112]\n"
     ]
    }
   ],
   "source": [
    "#appending values from columns into arrays\n",
    "ASP=data['AgeSpecificRate'].values\n",
    "CC=data['CaseCount'].values\n",
    "print(ASP)\n",
    "print(CC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "xarr=np.array(ASP)\n",
    "yarr=np.array(CC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "88.7836631848 25015.3641216\n"
     ]
    }
   ],
   "source": [
    "#calculating the regression line\n",
    "#intercept and slope of the line \n",
    "asp_mean=np.mean(ASP)\n",
    "cc_mean=np.mean(CC)\n",
    "m=len(ASP)\n",
    "j=0\n",
    "k=0\n",
    "for i in range(m):\n",
    "    j+=(ASP[i]-asp_mean)*(CC[i]-cc_mean)\n",
    "    k+=(ASP[i]-asp_mean)**2\n",
    "    \n",
    "#slope of the line\n",
    "slope=j/k \n",
    "\n",
    "#intercept\n",
    "intercept=cc_mean - (slope*asp_mean) \n",
    "\n",
    "print (slope,intercept)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "equation of best  fit line= 25015.3641216 + 88.7836631848 x\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKkAAAJQCAYAAABFDjdVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAwwmpAAAAABjuE49T9oopgtmNAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2d5665fb630>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the values and the fit line together\n",
    "#for the plot\n",
    "xlim=np.max(ASP)+10\n",
    "ylim=np.min(ASP)-10\n",
    "x=np.linspace(ylim,xlim,1000)\n",
    "\n",
    "#equation of line\n",
    "y=intercept+slope*x\n",
    "#Y=intercept+slope*X\n",
    "print(\"equation of best  fit line=\",intercept,\"+\",slope,\"x\")\n",
    "plt.plot(x,y,color='m', label='regression line')\n",
    "plt.scatter(xarr,yarr,c='b',label='scatter plot')\n",
    "\n",
    "plt.xlabel('AgeSpecificRate')\n",
    "plt.ylabel('CaseCount')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r_square value= 0.668377936996\n"
     ]
    }
   ],
   "source": [
    "############################TESTING###################################\n",
    "#r square method to determine the fit of the line to the data\n",
    "sum_sq_total =0\n",
    "sum_sq_res =0\n",
    "\n",
    "for i in range(m):\n",
    "    pred_eq = intercept+slope*xarr[i]\n",
    "    sum_sq_total+=(yarr[i]-cc_mean)**2\n",
    "    sum_sq_res+=(yarr[i]-pred_eq)**2\n",
    "    \n",
    "r_square=1-(sum_sq_res/sum_sq_total)\n",
    "\n",
    "print(\"r_square value=\",r_square)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#verification trough scipy to check the value of r is correct\n",
    "from scipy import stats\n",
    "\n",
    "slope, intercept, r_value, p_value, std_err = stats.linregress(xarr, yarr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "reverified value for r square= 0.668377936996\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "re_verif_value = r_value ** 2\n",
    "print (\"reverified value for r square=\",re_verif_value)\n",
    "np.isclose(r_square, re_verif_value, rtol=1e-15, atol=0.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "######################################################################\n",
      " 2nd Regression\n"
     ]
    }
   ],
   "source": [
    "print(\"######################################################################\\n 2nd Regression\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "######################################################################################## \n",
      "Performing Linear regression on dataset 2\n",
      "More about the dataset:\n",
      "From paper 'Composition-dependence of stacking fault energy in austenitic stainless steels through linear regression with random intercepts' \n",
      "(231, 2)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>C</th>\n",
       "      <th>SFE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.063</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.071</td>\n",
       "      <td>17.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.057</td>\n",
       "      <td>20.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.090</td>\n",
       "      <td>30.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.040</td>\n",
       "      <td>16.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       C   SFE\n",
       "0  0.063  14.0\n",
       "1  0.071  17.0\n",
       "2  0.057  20.0\n",
       "3  0.090  30.0\n",
       "4  0.040  16.0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plt.rcParams['figure.figsize']=(20.0,10.0)\n",
    "data = pd.read_csv('sfe.csv')\n",
    "#to have a heads up about the data we are dealing with\n",
    "\n",
    "print (\"######################################################################################## \\nPerforming Linear regression on dataset 2\")\n",
    "print (\"More about the dataset:\\nFrom paper 'Composition-dependence of stacking fault energy in austenitic stainless steels through linear regression with random intercepts' \")\n",
    "print(data.shape)\n",
    "#look at the first five rows of data for easier programming\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#appending values from columns into arrays\n",
    "C_content_wt=data['C'].values\n",
    "SFE=data['SFE'].values\n",
    "#print(C_content_wt) \n",
    "#print(SFE)\n",
    "\n",
    "xarr2=np.array(C_content_wt)\n",
    "yarr2=np.array(SFE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-14.8227523177 34.7533148714\n"
     ]
    }
   ],
   "source": [
    "#calculating the regression line\n",
    "#intercept and slope of the line \n",
    "c_cont_mean=np.mean(C_content_wt)\n",
    "SFE_mean=np.mean(SFE)\n",
    "mm=len(C_content_wt)\n",
    "jj=0\n",
    "kk=0\n",
    "for i in range(mm):\n",
    "    jj+=(C_content_wt[i]-c_cont_mean)*(SFE[i]-SFE_mean)\n",
    "    kk+=(C_content_wt[i]-c_cont_mean)**2\n",
    "    \n",
    "#slope of the line\n",
    "slope_2=jj/kk \n",
    "\n",
    "#intercept\n",
    "intercept_2=SFE_mean - (slope_2*c_cont_mean) \n",
    "\n",
    "print (slope_2,intercept_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "equation of best  fit line= 34.7533148714 + -14.8227523177 x\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABI8AAAJRCAYAAADMEZdnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAgAAAAAAQCTCIwAAAAAAAET6/59RXEHScF5+AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2d5685ac0b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the values and the fit line together\n",
    "#for the plot\n",
    "xlim2=np.max(C_content_wt)+0.1\n",
    "ylim2=np.min(C_content_wt)-0.1\n",
    "x2=np.linspace(ylim2,xlim2,1000)\n",
    "\n",
    "#equation of line\n",
    "y2=intercept_2+slope_2*x2\n",
    "\n",
    "print(\"equation of best  fit line=\",intercept_2,\"+\",slope_2,\"x\")\n",
    "plt.plot(x2,y2,color='m', label='regression line')\n",
    "plt.scatter(xarr2,yarr2,c='b',label='scatter plot')\n",
    "\n",
    "plt.xlabel('C_content_wt')\n",
    "plt.ylabel('SFE')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r_square value for c= 0.0410479885356\n"
     ]
    }
   ],
   "source": [
    "############################TESTING###################################\n",
    "#r square method to determine the fit of the line to the data\n",
    "sum_sq_total2 =0\n",
    "sum_sq_res2 =0\n",
    "\n",
    "for i in range(mm):\n",
    "    pred_eq2 = intercept_2+slope_2*xarr2[i]\n",
    "    sum_sq_total2+=(yarr2[i]-SFE_mean)**2\n",
    "    sum_sq_res2+=(yarr2[i]-pred_eq2)**2\n",
    "    \n",
    "r_square2=1-(sum_sq_res2/sum_sq_total2)\n",
    "\n",
    "print(\"r_square value for c=\",r_square2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "reverified value for r square for c= 0.0410479885356\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#verification trough scipy to check the value of r is correct\n",
    "\n",
    "slope, intercept, r_value, p_value, std_err = stats.linregress(xarr2, yarr2)\n",
    "re_verif_value2 = r_value ** 2\n",
    "print (\"reverified value for r square for c=\",re_verif_value2)\n",
    "np.isclose(r_square2, re_verif_value2, rtol=1e-13, atol=0.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "we will check how other components in the alloy and SFE are corelated\n",
      "######################################################################################## \n",
      "Performing Linear regression on dataset 2\n",
      "More about the dataset:\n",
      "From paper 'Composition-dependence of stacking fault energy in austenitic stainless steels through linear regression with random intercepts' \n",
      "(223, 3)\n",
      "0.71957470718 23.7905725502\n",
      "equation of best  fit line= 23.7905725502 + 0.71957470718 x\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABI8AAAJRCAYAAADMEZdnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAMYSHgEAAAAw1v8Dv6phI7HkfbMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2d569149668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r_square value for ni= 0.0829532642716\n",
      "reverified value for r square for ni= 0.0829532642716\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "##########################################################################\n",
    "print(\"we will check how other components in the alloy and SFE are corelated\")\n",
    "#####################\n",
    "#rehashing same code from above\n",
    "####################\n",
    "plt.rcParams['figure.figsize']=(20.0,10.0)\n",
    "data = pd.read_csv('sfe_ni_cr.csv')\n",
    "#to have a heads up about the data we are dealing with\n",
    "\n",
    "print (\"######################################################################################## \\nPerforming Linear regression on dataset 2\")\n",
    "print (\"More about the dataset:\\nFrom paper 'Composition-dependence of stacking fault energy in austenitic stainless steels through linear regression with random intercepts' \")\n",
    "print(data.shape)\n",
    "#look at the first five rows of data for easier programming\n",
    "data.head()\n",
    "\n",
    "#appending values from columns into arrays\n",
    "ni_content_wt=data['Ni'].values\n",
    "SFE=data['SFE'].values\n",
    "#print(ni_content_wt) \n",
    "#print(SFE)\n",
    "\n",
    "xarr2a=np.array(ni_content_wt)\n",
    "yarr2a=np.array(SFE)\n",
    "\n",
    "#calculating the regression line\n",
    "#intercept and slope of the line \n",
    "ni_cont_mean=np.mean(ni_content_wt)\n",
    "SFE_mean=np.mean(SFE)\n",
    "mm2=len(ni_content_wt)\n",
    "jj2=0\n",
    "kk2=0\n",
    "for i in range(mm2):\n",
    "    jj2+=(ni_content_wt[i]-ni_cont_mean)*(SFE[i]-SFE_mean)\n",
    "    kk2+=(ni_content_wt[i]-ni_cont_mean)**2\n",
    "    \n",
    "#slope of the line\n",
    "slope_2a=jj2/kk2 \n",
    "\n",
    "#intercept\n",
    "intercept_2a=SFE_mean - (slope_2a*ni_cont_mean) \n",
    "\n",
    "print (slope_2a,intercept_2a)\n",
    "\n",
    "#plotting the values and the fit line together\n",
    "#for the plot\n",
    "xlim2a=np.max(ni_content_wt)+5\n",
    "ylim2a=np.min(ni_content_wt)-5\n",
    "x2a=np.linspace(ylim2a,xlim2a,1000)\n",
    "\n",
    "#equation of line\n",
    "y2a=intercept_2a+slope_2a*x2\n",
    "\n",
    "print(\"equation of best  fit line=\",intercept_2a,\"+\",slope_2a,\"x\")\n",
    "plt.plot(x2a,y2a,color='m', label='regression line')\n",
    "plt.scatter(xarr2a,yarr2a,c='b',label='scatter plot')\n",
    "\n",
    "plt.xlabel('ni_content_wt')\n",
    "plt.ylabel('SFE')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "############################TESTING###################################\n",
    "#r square method to determine the fit of the line to the data\n",
    "sum_sq_total2a =0\n",
    "sum_sq_res2a =0\n",
    "\n",
    "for i in range(mm2):\n",
    "    pred_eq2a = intercept_2a+slope_2a*xarr2a[i]\n",
    "    sum_sq_total2a+=(yarr2a[i]-SFE_mean)**2\n",
    "    sum_sq_res2a+=(yarr2a[i]-pred_eq2a)**2\n",
    "    \n",
    "r_square2a=1-(sum_sq_res2a/sum_sq_total2a)\n",
    "\n",
    "print(\"r_square value for ni=\",r_square2a)\n",
    "\n",
    "#verification trough scipy to check the value of r is correct\n",
    "\n",
    "slope, intercept, r_value, p_value, std_err = stats.linregress(xarr2a, yarr2a)\n",
    "re_verif_value2a = r_value ** 2\n",
    "print (\"reverified value for r square for ni=\",re_verif_value2a)\n",
    "np.isclose(r_square2a, re_verif_value2a, rtol=1e-13, atol=0.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "we will check how other components in the alloy and SFE are corelated\n",
      "######################################################################################## \n",
      "Performing Linear regression on dataset 2\n",
      "More about the dataset:\n",
      "From paper 'Composition-dependence of stacking fault energy in austenitic stainless steels through linear regression with random intercepts' \n",
      "(223, 3)\n",
      "0.659311050035 24.6298885547\n",
      "equation of best  fit line= 24.6298885547 + 0.659311050035 x\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABI8AAAJQCAYAAAAHTUTCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3X2UZWV9J/rvru5GKFAjLXGRZaoLM0pMVFptGJVMfAEl\ny7iMyaiJU2RQkrQBlzfRca6Y0rzMshMSneRmRSGrHFGiJRfH5EbHMBOFuUZjXoYGMRoUmXG6+qKo\n0CYolEA3ve8fu4vuqj67+5yq87L3OZ/PWr1O7V3n7P3svc+p6v2t5/k9RVmWAQAAAIBOpkbdAAAA\nAACaS3gEAAAAQC3hEQAAAAC1hEcAAAAA1BIeAQAAAFBLeAQAAABALeERAAAAALWERwAAAADUEh4B\nAAAAUGvzqBvQjcc+9rHl7OzsqJsBAAAAMDZuuummu8uyPO14z2tFeDQ7O5vdu3ePuhkAAAAAY6Mo\niqVunmfYGgAAAAC1hEcAAAAA1BIeAQAAAFCrFTWPAAAAgGbYv39/7rjjjtx///2jbgpdOvHEE/P4\nxz8+W7ZsWdfrhUcAAABA1+6444488pGPzOzsbIqiGHVzOI6yLLNv377ccccdOeOMM9a1DcPWAAAA\ngK7df//92bp1q+CoJYqiyNatWzfUU0x4BAAAAPREcNQuG71ewiMAAAAAagmPAAAAADbo13/913P9\n9ddveDuf+tSn8pKXvCRJ8rGPfSyXX375hre5UQpmAwAAAK1UlmXKsszU1Pr7xhw4cCCbN288HvkP\n/+E/bHgba730pS/NS1/60r5vt1fCIwAAAGBdbv/V23PvLff2dZunbD8lT/y/nlj7/T179uSCCy7I\nv/yX/zI33XRTrrvuutx22235jd/4jTzwwAP5oR/6obzvfe/LKaeckuuuuy5vfOMbc/LJJ+fcc8/N\nV7/61Xz84x/Pb/7mb+Z//a//la9+9auZmZnJBz/4wVx22WX51Kc+lQceeCCve93r8trXvjZ33nln\nfvZnfzbf+c53cuDAgVx55ZV5znOek1/4hV/I7t27UxRFLr744rzhDW/Iq1/96rzkJS/Jy1/+8txw\nww1505velAMHDuTss8/OlVdemUc84hGZnZ3NRRddlP/yX/5L9u/fn//8n/9zfviHf7j2WN///vdn\n9+7dede73pVXv/rVedSjHpXdu3fnG9/4Rn7v934vL3/5y5Mk73jHO/LhD384DzzwQH76p386v/Vb\nv9XXa2LYGgAAANAqt99+ey699NL84z/+Y04++eS8/e1vz/XXX5+bb745O3bsyO///u/n/vvvz2tf\n+9r81//6X3PTTTflrrvuWrWNW2+9Nddff32uueaavPe9782jH/3o3Hjjjbnxxhvznve8J//7f//v\nfOhDH8oFF1yQW265JZ///Oezffv23HLLLfna176WL37xi/nCF76Q17zmNau2e//99+fVr351rr32\n2nzhC194OHRa8djHPjY333xzLrnkkrzzne/s6bjvvPPO/PVf/3U+/vGP57LLLkuSfOITn8jtt9+e\n//E//kduueWW3HTTTfn0pz+9zjPbmZ5HAAAAwLocq4fQIG3bti3PetazkiR/93d/l1tvvTXnnntu\nkuTBBx/Ms5/97Hz5y1/OE57whJxxxhlJkle96lVZWFh4eBsvfelLc9JJJyWpAph/+Id/yEc+8pEk\nyT333JPbb789Z599di6++OLs378/L3vZy7J9+/Y84QlPyFe/+tW8/vWvz0/+5E/mRS960aq23Xbb\nbTnjjDPypCc9KUly0UUX5d3vfnd+9Vd/NUnyMz/zM0mSZz7zmfmzP/uzno77ZS97WaampvIjP/Ij\n+eY3v/lw2z/xiU/k6U9/epLk3nvvze23354f//Ef72nbxyI8AgAAAFrl5JNPfvjrsizzwhe+MNdc\nc82q59xyyy09beOP/uiPcsEFFxz1vE9/+tP5i7/4i7z61a/OG9/4xvzbf/tv8/nPfz5/+Zd/mT/+\n4z/Ohz/84Vx11VVdt/0Rj3hEkmTTpk05cOBA16878rUrbV55fMtb3pLXvva1PW2rF4atAQAAAK31\nrGc9K5/97GfzP//n/0yS3HffffnKV76SM888M1/96lezZ8+eJMm1115bu40LLrggV155Zfbv358k\n+cpXvpL77rsvS0tLedzjHpdf+qVfyi/+4i/m5ptvzt13352DBw/mX//rf523v/3tufnmm1dt68wz\nz8yePXsebs8HPvCBPPe5zx3AkR9u+1VXXZV7761qT33ta1/Lt771rb7uQ88jAAAAoLVOO+20vP/9\n78+rXvWqPPDAA0mSt7/97XnSk56UK664Ij/xEz+Rk08+OWeffXbtNn7xF38xe/bsyTOe8YyUZZnT\nTjstf/7nf55PfepTecc73pEtW7bklFNOyZ/8yZ/ka1/7Wl7zmtfk4MGDSZLf+Z3fWbWtE088Me97\n3/vyile84uGC2b/8y788sON/0YtelC996Ut59rOfnSQ55ZRT8sEPfjDf//3f37d9FCvdnJpsx44d\n5e7du0fdDAAAAJh4X/rSl/LkJz951M3oyr333ptTTjklZVnmda97XZ74xCfmDW94w6ibNRKdrltR\nFDeVZbnjeK81bA0AAAAYS+95z3uyffv2/OiP/mjuueeegdYFGmeGrQEAAABj6Q1veMPE9jTqJz2P\nAAAAAKglPAIAAOhgcTGZnU2mpqrHxcVRtwhgNAxbAwAAWGNxMdm5M1lerpaXlqrlJJmbG127AEZB\nzyMAAIA15ucPB0crlper9QCTRngEAACwxt69va0H2mXPnj350Ic+9PDyLbfckuuuu25g+3ve856X\n3bt3H/M5v/3bvz2w/W+U8AgAAGCNmZne1gP1mlg/rB/h0YEDB/raJuERAABAi+zalUxPr143PV2t\nB7q3Uj9saSkpy8P1wzYSIN133335yZ/8yZx11ll5ylOekmuvvTZJcuONN+Y5z3lOzjrrrJxzzjn5\n7ne/mz179uRf/at/lWc84xl5xjOekb/5m79Jklx22WX5zGc+k+3bt+d3f/d38+u//uu59tprs337\n9lx77bW57777cvHFF+ecc87J05/+9Hz0ox9Nkrz//e/PS1/60rzgBS/Ieeedt6pde/bsyQ//8A9n\nbm4uT37yk/Pyl788y2vHvya55ppr8tSnPjVPecpT8uY3v/nh9nzve9/L9u3bM9fAwmoKZgMAAKyx\ncu82P18NVZuZqYKjBt7TQaMdq37Yej9P/+2//bf8wA/8QP7iL/4iSXLPPffkwQcfzM/+7M/m2muv\nzdlnn53vfOc7Oemkk/L93//9+eQnP5kTTzwxt99+e171qldl9+7dufzyy/POd74zH//4x5Mkj3vc\n47J79+68613vSpL82q/9Wl7wghfkqquuyj//8z/nnHPOyfnnn58kufnmm/MP//APOfXUU49q2223\n3Zb3vve9Offcc3PxxRfniiuuyJve9KaHv//1r389b37zm3PTTTflMY95TF70ohflz//8z3P55Zfn\nXe96V2655Zb1nZQB0/MIAACgg7m5ZM+e5ODB6lFwBL0bRP2wpz71qfnkJz+ZN7/5zfnMZz6TRz/6\n0bntttty+umn5+yzz06SPOpRj8rmzZuzf//+/NIv/VKe+tSn5hWveEVuvfXWrvbxiU98Ipdffnm2\nb9+e5z3vebn//vuz91CjX/jCF3YMjpLkB3/wB3PuuecmSS688ML89V//9arv33jjjXne856X0047\nLZs3b87c3Fw+/elPr/dUDI2eRwAAAMBAzMxUQ9U6rV+vJz3pSbn55ptz3XXX5a1vfWvOO++8/PRP\n/3TH5/7BH/xBHve4x+Xzn/98Dh48mBNPPLGrfZRlmT/90z/NmWeeuWr93//93+fkk0+ufV1RFMdc\nbis9jwAAAICBGET9sK9//euZnp7OhRdemH//7/99br755px55pm58847c+ONNyZJvvvd7+bAgQO5\n5557cvrpp2dqaiof+MAH8tBDDyVJHvnIR+a73/3uw9tcu3zBBRfkj/7oj1KWZZLkc5/7XFdt27t3\nb/72b/82SfKhD30oP/ZjP7bq++ecc07+6q/+KnfffXceeuihXHPNNXnuc5+bJNmyZUv279+/zrMy\nWMIjAAAAYCDm5pKFhWTbtqQoqseFhY0NA/3CF76Qc845J9u3b89v/dZv5a1vfWtOOOGEXHvttXn9\n61+fs846Ky984Qtz//3359JLL83VV1+ds846K1/+8pcf7jX0tKc9LZs2bcpZZ52VP/iDP8jzn//8\n3HrrrQ8XzH7b296W/fv352lPe1p+9Ed/NG9729u6atuZZ56Zd7/73Xnyk5+cf/qnf8oll1yy6vun\nn356Lr/88jz/+c/PWWedlWc+85n5qZ/6qSTJzp0787SnPa2RBbOLlRStyXbs2FHu3r171M0AAACA\nifelL30pT37yk0fdjMbZs2dPXvKSl+SLX/ziqJvSUafrVhTFTWVZ7jjea/U8AgAAAKCW8AgAAABg\ng2ZnZxvb62ijhEcAAABAT9pQAofDNnq9hEcAAABA10488cTs27dPgNQSZVlm3759OfHEE9e9jc19\nbA8AAAAw5h7/+MfnjjvuyF133TXqptClE088MY9//OPX/XrhEQAAANC1LVu25Iwzzhh1Mxgiw9YA\nAAAAqCU8AgAAAKCW8AgAAACAWsIjAAAAAGoJjwAAAACoJTwCAAAAoJbwCAAAAIBawiMAAAAAagmP\nAAAAAKglPAIAAACglvAIAAAAgFrCIwAAAABqCY8AAAAAqCU8AgAAAKCW8AgAAACAWsIjAAAAAGoJ\njwAAAACoJTwCAAAAoJbwCAAAAIBawiMAAAAAagmPAAAAAKglPAIAAACglvAIAAAAgFrCIwAAAABq\nCY8AAAAAqCU8AgAAAKCW8AgAAACAWsIjAAAAAGoJjwAAAACoJTwCAAAAoJbwCAAAAIBawiMAAAAA\nagmPAAAAAKglPAIAAACglvAIAAAAgFrCIwAAAABqCY8AAAAAqCU8AgAAAKCW8AgAAACAWsIjAAAA\nAGoJjwAAAACoJTwCAAAAoJbwCAAAAIBawiMAAAAAagmPAAAAAKglPAIAAACglvAIAAAAgFrCIwAA\nAABqCY8AAAAAqCU8AgAAAKCW8AgAAACAWsIjAAAAAGoJjwAAAACoJTwCAAAAoJbwCAAAAIBawiMA\nAAAAagmPAAAAAKglPAIAAACg1kDDo6Io3lIUxa1FUXyxKIpriqI4sSiKU4ui+GRRFLcfenzMINsA\nAAAAwPoNLDwqimI2yc4kzyzL8ilJNiX5uSSXJbmhLMsnJrnh0DIAAAAADTTInkffSbI/yUlFUWxO\nMp3k60l+KsnVh55zdZKXDbANAAAAAGzAwMKjsiy/neSdSfYmuTPJPWVZfiLJ48qyvPPQ076R5HGd\nXl8Uxc6iKHYXRbH7rrvuGlQzAQAAADiGQQ5b+6Ekb0hyRpIfSHJyURQXHvmcsizLJGWn15dluVCW\n5Y6yLHecdtppg2omAAAAAMcwyGFrO5L8TVmWd5VluT/JnyV5TpJvFkVxepIcevzWANsAAAAAwAYM\nMjy6LcmziqKYLoqiSHJeki8l+ViSiw4956IkHx1gGwAAAADYgM2D2nBZlrcURfEnSXYnOZjkc0kW\nkpyS5MNFUfxCkqUkrxxUGwAAAADYmIGFR0lSluXvJvndNasfSNULCQAAAICGG+SwNQAAAABaTngE\nAAAAQC3hEQAAtRYXk9nZZGqqelxcHHWLAIBhG2jNIwAA2mtxMdm5M1lerpaXlqrlJJmbG127AIDh\n0vMIAICO5ucPB0crlper9QDA5BAeAQDQ0d69va0HAMaT8AgAgI5mZnpbDwCMJ+ERAAAd7dqVTE+v\nXjc9Xa0HACaH8AgAgI7m5pKFhWTbtqQoqseFBcWyAWDSmG0NAIBac3PCIgCYdHoeAQAAAFBLeAQA\nAABALeERAAAAALWERwAAAADUEh4BAAAAUEt4BAAAAEAt4REAAAAAtYRHAAAAANQSHgEAAABQS3gE\nAAAAQC3hEQAAAAC1hEcAAAAA1BIeAQAAAFBLeAQAAABALeERAAAAALWERwAAAADUEh4BAAAAUEt4\nBAAAAEAt4REAAAAAtYRHAAA03uJiMjubTE1Vj4uLo24RAEyOzaNuAAAAHMviYrJzZ7K8XC0vLVXL\nSTI3N7p2AcCk0PMIAIBGm58/HBytWF6u1gMAgyc8AgCg0fbu7W09ANBfwiMAABptZqa39QBAfwmP\nAABotF27kunp1eump6v1AMDgCY8AAGi0ublkYSHZti0piupxYUGxbAAYFrOtAQDQeHNzwiIAGBU9\njwAAAACoJTwCAAAAoJbwCAAAAIBawiMAAAAAagmPAAAAAKglPAIAAACglvAIAAAAgFrCIwAAAABq\nCY8AAAAAqCU8AgAAAKCW8AgAAACAWsIjAAAAAGoJjwAAAACoJTwCAAAAoJbwCAAAAIBawiMAAAAA\nagmPAAAAAKglPAIAAACglvAIAAAAgFrCIwAAura4mMzOJlNT1ePi4qhbBAAM2uZRNwAAgHZYXEx2\n7kyWl6vlpaVqOUnm5kbXLgBgsPQ8AgCgK/Pzh4OjFcvL1XoAYHwJjwAA6Mrevb2tBwDGg/AIAICu\nzMz0th4AGA/CIwAAurJrVzI9vXrd9HS1HgAYX8IjAAC6MjeXLCwk27YlRVE9Liwolg0A485sawAA\ndG1uTlgEAJNGzyMAAAAAagmPAAAAAKglPAIAAACglvAIAAAYmsXFZHY2mZqqHhcXR90iAI5HwWwA\nAGAoFheTnTuT5eVqeWmpWk4UYgdoMj2PAACAoZifPxwcrVhertYD0FzCIwAAYCj27u1tPQDNIDwC\nAACGYmamt/UANIPwCACAxpv0Isvjcvy7diXT06vXTU9X6wFoLgWzAQBotEkvsjxOx7/S3vn5aqja\nzEwVHLXtOAAmTVGW5ajbcFw7duwod+/ePepmAAAwArOzVWCy1rZtyZ49w27N8E368QMwOEVR3FSW\n5Y7jPc+wNQAAGm3SiyxP+vEDMHrCIwAAGm3SiyxP+vEDMHrCIwAAGm3SiyxP+vEDMHrCIwAAGm1u\nLllYqGr8FEX1uLAwOUWWN3r84zJT26Ry/YAmUDAbAADG1NqZ2pKq19IkhW9t5voBg9ZtwWzhEQAA\njCkztbWb6wcMmtnWAABgwpmprd1cP6AphEcAADCmzNTWbq4f0BTCIwAAGFNmams31w9oCuERAACM\nqUmfqa7tXD+gKRTMBgAAAJhACmYDAAAAsGHCIwAAAABqCY8AAAAAqCU8AgCg8RYXk9nZZGqqelxc\nHHWLAGBybB51AwAA4FgWF5OdO5Pl5Wp5aalaTsw6BQDDoOcRAACNNj9/ODhasbxcrQcABk94BABA\no+3d29t6AKC/hEcAADTazExv6wGA/hIeAQDQaLt2JdPTq9dNT1frx4mi4AA0lfAIAIBGm5tLFhaS\nbduSoqgeFxbGq1j2SlHwpaWkLA8XBRcgAdAERVmWo27Dce3YsaPcvXv3qJsBAAADMTtbBUZrbduW\n7Nkz7NYAMCmKoripLMsdx3uenkcAADBiioID0GTCIwAAGDFFwQFoMuERAACM2KQUBQegnQYaHhVF\n8X1FUXykKIovF0XxpaIonl0UxalFUXyyKIrbDz0+ZpBtABgFM+YA0ItJKAoOQHsNtGB2URRXJ/lM\nWZb/qSiKE5JMJ/m1JN8uy/LyoiguS/KYsizffKztKJgNtMnKjDnLy4fXTU+7CQAAAJql24LZAwuP\niqJ4dJJbkjyhPGInRVHcluR5ZVneWRTF6Uk+VZblmcfalvAIaBMz5gAAAG3QhNnWzkhyV5L3FUXx\nuaIo/lNRFCcneVxZlncees43kjyu04uLothZFMXuoih233XXXQNsJkB/mTEHAAAYJ4MMjzYneUaS\nK8uyfHqS+5JcduQTDvVI6tj1qSzLhbIsd5RlueO0004bYDMB+suMOQAAwDgZZHh0R5I7yrL8+0PL\nH0kVJn3z0HC1HHr81gDbADB0ZswBAADGycDCo7Isv5Hk/yuKYqWe0XlJbk3ysSQXHVp3UZKPDqoN\nAKNgxhwAAGCcDHq2te1J/lOSE5J8NclrUgVWH04yk2QpySvLsvz2sbajYDYAAABAf3VbMHvzIBtR\nluUtSTo14rxB7hcAAACA/hhkzSMAAAAAWk54BAAAAEAt4REAAAAAtYRHAAAAANQSHgEAAABQS3gE\nAAAAQC3hEQAAAAC1hEcAAADQpcXFZHY2mZqqHhcXR90iGLzNo24AAAAAtMHiYrJzZ7K8XC0vLVXL\nSTI3N7p2waDpeQQAAABdmJ8/HBytWF6u1sM4Ex4BAABAF/bu7W09jAvhEQAAAHRhZqa39TAuhEcA\nAADQhV27kunp1eump6v1MM6ERwAAANCFublkYSHZti0piupxYUGxbMaf2dYAAACgS3NzwiImj55H\nAAAAR1hcTGZnk6mp6nFxcdQtAhgtPY8AAAAOWVxMdu48PB370lK1nOhtAkwuPY8AAAAOmZ8/HByt\nWF6u1gNMKuERAADAIXv39rYeYBIIjwAAAA6ZmeltPcAkEB4BAAAcsmtXMj29et30dLUeYFIJjwAA\nAA6Zm0sWFpJt25KiqB4XFhTLBiab2dYAAACOMDcnLAI4kp5HAAAAANQSHgEAAABQS3gEAAAAQC3h\nEQAAAAC1hEcAAAAA1BIeAQAAY2txMZmdTaamqsfFxXbuA2CUhEcA0EduIGA0fPboZHEx2bkzWVpK\nyrJ63Lmzv++PYewDYNSKsixH3Ybj2rFjR7l79+5RNwMAjmnlBmJ5+fC66elkYSGZmxtdu2Dc+exR\nZ3a2CnPW2rYt2bOnPfsAGJSiKG4qy3LHcZ8nPAKA/nADAaPhs0edqamqN9BaRZEcPNiefQAMSrfh\nkWFrANAne/f2th7oD5896szM9La+qfsAGDXhEQD0iRsIGA2fPers2lUNYTzS9HS1vk37ABg14REA\n9IkbCMZRUwpRH6sdPnvUmZural9t21YNI9u2rf+1sIaxD4BRU/MIAPpocTGZn6+Gy8zMVDevbiBo\nq6YUou6mHT57ANA7BbMBANiQphSibko7AGDcKJgNAMCGNKUQdVPaAQCTSngEAEBHTSlE3ZR2AMCk\nEh4BANBRUwpRN6UdADCphEcAAHTUlFmkmtIOAJhUCmYDANA3Zj0DgPbotmD25mE0BgCA8be4mOzc\nmSwvV8tLS9VyIkACgDYzbA0AgL6Ynz8cHK1YXq7W11lcTGZnk6mp6nFxcZAtBADWQ88jAAD6Yu/e\n3tbrqQQA7aDnEQyZv7BCf/gsQfPMzPS2fj09lWgnP7MB2k14BEO08hfWpaWkLA//hdV/oKA3PkvQ\nTLt2JdPTR6+/997On89eeyrRTn5mA7Sf8AiGyF9YoT98lmAw1ts7ZOV1P//zyUknJSefvPr7+/Z1\nDgt67alEO/mZDdB+wiMYIn9hhf7wWYL+W0/vkMXF5LGPTS688PDr9u07OihIOocFnXoqTU9X6xkf\nfmYDtJ/wCIbIX1ihP3yWoP967R2yEjbt23f098qy82vWhgVzc8nCQrJtW1IU1ePCgmLZ48bPbID2\nO2Z4VBTF/3nE169Y873fHlSjYFz5Cyv0h88S9F+vvUM6hU3H0yksmJtL9uxJDh6sHgVH48fPbID2\nO17Po5874uu3rPneT/S5LTD2/IUV+sNnCY6v1/pFdb1ApqZ6K3ZdZ8sWYcGk8jMboP2Ksq5fcZKi\nKD5XluXT137daXmQduzYUe7evXsYuwIAaL2VIWVH9gyanj72DXun1xzrtbOzVZ2jTrZsqYauHThw\neN0JJyRXXSUwAIAmKYriprIsdxzvecfreVTWfN1pGQCABljP7FYrvUM2bTr6e90Wu06SrVuTRz1q\ndXCUJA8+aHYtAGir44VHZxVF8Z2iKL6b5GmHvl5ZfuoQ2gcAQI/WO7vV3FxVe6ib13YaivTBDyZ3\n3518+9vr2z/jodchkwA03+ZjfbMsyw5/ewIAoMlmZjoPKetmdqteXjs313kY2kb2T7utHf64tFQt\nJ4YsArTZ8WZb+5kjvn7M4JsDAO3mL+40wUZmt+rHzFhm15pc6xkyCUDzHW/Y2luP+PqGQTYEANpu\n5S/uS0tVseCVv7gLkBi2Xme3OjL0nJ9PLrpoYzNjmV1rcq13yCQAzbbu2daGyWxrALRB3exT27Yl\ne/YMuzXQnfXMzAZ1/BwEaJd+zbZ2UlEUTy+K4plJTiyK4hlH/utPUwFgPPiLO21kmBH9NM5DFg1L\nBibZMQtmJ7kzyX9MUiT5RpJ3rvn+CwbRKABoI0WCaSOhJ/200lttfr56D83MVMFR23uxKQQOTLrj\n9Tx6c5K5siyfX5bl85NcneTeJF9M8vJBNw4A2mSc/+LO+KoLNwcdeo5TL45xOpZ+mJurhqgdPFg9\njkO4UtdD71d+xbUHJsPxwqM/TvJAkhRF8eNJfidVgHRPkoXBNg0A2kWRYNpoFKHnOBWXH6djoV5d\nT7x9+1x7YDIcr2D258uyPOvQ1+9OcldZlr95aPmWsiy3D6ORCmYDwPhbXBy/oS5tMexzP05Flcfp\nWKhXd507ce2BNulXwexNRVGs1EU6L8l/P+J7x6uXBADQFb03RqtumNGghmONU52lcToW6nXqoVfH\ntQfG0fHCo2uS/FVRFB9N8r0kn0mSoij+RaqhawAAG2bGr2Y4Mix67GOT17xmMIHeqOosDcI4HQv1\nOg1L3rq183Nde2AcHTM8KstyV5J/l+T9SX6sPDzGbSrJ6wfbNABgUui9MXpre3/t25fs37/6Of0K\n9MapuPw4HQvHtraH3h/+oWsPTI7j9TxKWZZ/V5bl/1OW5X1HrPtKWZY3D7ZpAMCkaErvjUmeNatT\n769O+hHo9aO4fFOulUL5k8u1BybJMQtmN4WC2QAw3lZ6vRwZXkxPD/dGrAltGKWpqarH0fE0oRjw\npF8rAOiXfhXMBgAYuCb8BX/S6y5108urKUNyJv1aAcCwCY8AgEaom/FrWCa97lKn2j0nnFAVBW7a\nkJymXqtrZ9V4AAAgAElEQVSmDKUDgH4THgEApDl1l0alU++vq65K7r57dIFenSZeq7UFx/s5Ox0A\njJrwCAAgZs1KBtf7q989cpp4rQylm0x6mwGTQngEAJBm1F0aR4PokdPEa9XUoXQMjt5mwCQx2xoA\nAAMzO1vdVK/VhFnb+mlSjpPDXHNgHJhtDQCAkZuUHjlNHErHYE3KexsgER4BADBATSxuPQhNHErH\nYE3KexsgER4BADBAk9QjZ1AFx2mmSXpvAwiPAAD6wKxLnemRw7jy3gYmiYLZAAAbtDLr0pFTtU9P\nu5EEAJpNwWwAgCGZn18dHCXV8vz8aNoDANBPwiMAoNHaMBzMrEsAwDgTHgEAjbUyHGxpKSnL6nHn\nzuYFSGZdAgDGmfAIAGistgwHG/asS23ojdWLQR3PsM7TuF0PAFhLeAQANFZbhoMNc9altvTG6tag\njmdY52ncrge9ERwCk8JsawBAY83OVjfja23bluzZM+zWNMO4nZNBHc+wztO4XQ+6Z5ZFYBx0O9ua\n8AgAaCw3Z0ebmqp6uKxVFMnBg8Nvz0YN6niGdZ7G7XrQPcEhMA66DY8MWwMAGmuYw8HaYtyKc596\nauf1Gz2eYZ2ncbsedK8tw2oB+kF4BAA02txc9Vf8gwerx0kOjpLeinM3vR7L4mLy3e8evX7Llo0X\nGx9WEfNhF0unOQSHwCQRHgEw1pp+8wy9Ol5vrJX3fFEkF164upDza17T22dg0J+f+fnkwQePXv+o\nRx0dEvbalmH1WtM7bnIJDoFJouYRAGNLvRwmTaf3/FpbtyZ3372+bfX789NtvSCfZZpqcbEKQffu\nrXoc7drlPQm0i4LZAEw8xUyZNHXv+bW6+e/fMD4/3e7DZxkABkPBbAAmnmKmTJp+vreH8fnpdtiP\nzzIAjJbwCICxpZgpk6ab9/bWrRvbVj8/P93WC/JZBoDREh4BMLYUM2XSdHrPr/WHf7j+bQ3i89PN\nbHpt/iwr2g/AOBAeATC2zILEJDrppPrvFUVV3Pd4QcZKEeDl5WTTpmrdKD8/bf0srxT6PnLGu507\nBUgAtI+C2QAAY6CbmdamplbPYtZpxrK6mc0uuii57jqzSvVCoW8Ams5sawAAE6TbmdbW6nZms6JY\nPUtbp+CJ1aamOs9sVxSrQzwAGBWzrQEATJD1zjy29nV121kbgiwvV0PbqKfQNwDjYuDhUVEUm4qi\n+FxRFB8/tHxqURSfLIri9kOPjxl0GwAAxt16A4m1rzv11O5fu97AalK0udA3ABxpGD2PfiXJl45Y\nvizJDWVZPjHJDYeWAQBabdSzanUz01rd69ar3z1o1nMOR33ej9WGthb6niRNeP8cTxvaCIy/gdY8\nKori8UmuTrIryRvLsnxJURS3JXleWZZ3FkVxepJPlWV55rG2o+YRANBkdUWmhx0UrMyStndvdaP5\n0EPHfv7Wrcndd69+bbd1k/p9fOs5h004701oA+vThmvXhjYC7daIgtlFUXwkye8keWSSNx0Kj/65\nLMvvO/T9Isk/rSzXER4BAE3WxFm1FheTiy9OHnyw8/ePvAHtZqa2rVuTU04Z3Gxr6zmHTTjvTWgD\n69OGa9eGNgLt1m14tHmADXhJkm+VZXlTURTP6/ScsizLoig6pldFUexMsjNJZlQVBAAarK72z6hr\nAtX9jXDr1urx53++6m10773HDo6S5JWvTK64or/tO9J6zmETznsT2sD6tOHataGNwGQYZM2jc5O8\ntCiKPUn+7yQvKIrig0m+eWi4Wg49fqvTi8uyXCjLckdZljtOO+20ATYTAGBjmjir1vx8sn//0eu3\nbk2+971k374qXFpaqr4+noWFwdZaWc85bMJ5b0IbWJ82XLs2tBGYDAMLj8qyfEtZlo8vy3I2yc8l\n+e9lWV6Y5GNJLjr0tIuSfHRQbQAAGIYmzqpV1zNh377j9zLq5KGHqqFtgwqQjncOOxUNbsJ5b0Ib\nWJ82XLs2tBGYDMOYbW2ty5O8sCiK25Ocf2gZAKC1mjir1iB6JiwvVz2aBuFY53ClJtPS0uHeUjt3\nVq8b9Xlv4rWnO224dm1oIzAZBlowu18UzAYA6E3dLE0nndTdMLU6RZEcPLjx9vVC0WAAGIxuC2aP\noucRAAADVtdj4ZWv3Nh2R1FrRdFgABgt4REAtFCn+i+w1txc1TPn4MHqcW4uue66jW3zxS/uR8t6\no2jwcPi5AkAd4REAtExd/Rc3enRjo711Nho+rUfTiwaff37Vu2vl3/nnj7pFvfNzZX0EbsCkUPMI\nAFpG/Rc2ou79061R1DxKqpvy+fkq/JqZqYKjJhQNPv/85IYbjl5/3nnJ9dcPvz3r5edK7+rqiilo\nDbRJtzWPhEcA0DJTU1XPgLVGdVNPu3S64T3hhOo9tX//8V8vTFitKOq/14L/Zj/Mz5XeCdyAcaBg\nNgCMKfVf6KTb4TOdCmlfdVXyvvetXnfJJYMbKmaoT/P4udI7hdyBSSI8AoCWaXr9F4av13o1nQpp\nr113xRWdZ2vb6HActXWayc+V3gncgEkiPAKAlqmbgl2Njck1P796GFpSLc/Pb2y7nUKmjRpUW0fl\nvPN6W99Ufq70TuAGTBLhEQC00CBu6umfYQ/L2sjwmTa1tR/6fbzXX390UNS2Ytkr/FzpjcANmCQK\nZgMA9NEoZmBab+HeNrW1H8yOBQCrKZgNADACoxiWtd7hM21qaz+M25A5ABgW4REAQB+NYljWeofP\ntKmt/TDqIXMA0FbCIwCgFdoyvfuoZmBaT72aNrW1H8yOBQDrIzwCABqvTdO79zIsa9SBWKe2Jsm9\n9/bWllEfR7cGPWSuLecBAHolPAIAGq9NtWq6HZbVhEBspa1bt65ev29f921pwnF0a5BD5tp0HgCg\nV2ZbAwAab2qquiFfqyiqoU9tNMpZx/rZliYdxyg5DwC0kdnWAGAEDFsZjHGsVdOk4s2dQo+ku7Y0\n6ThGyXkAYJwJjwCgTwxbGZxRTu8+KE0JxBYXqx5c621LU45j1JwHAMaZ8AgA+qRNdXnaZpTTuw9K\nUwKx+fn6IYHdtKUpxzFqzgMA40x4BAB9YtjKYI1qevdBaUogVvf+LMvObVk7NDNpxnF0MsxhpE25\nnrSf4c9AEymYDQB9omAubdTL+3ZlaOaRPeymp5sZkrSprbDC+xYYNgWzAWDIDFuhjXp537ZpaGab\n2gorvG+BphIeAUCfGLZCG/Xyvm3L0MzFxY3NIAej0pbPGDB5DFsDAKArbRia2WnYz5Ga1FZYqw2f\nMWC8GLYGAEBf9Xto5iAKA3ca9rNiUMNIzz+/6rW18u/88/u/DyaD4c9AUwmPAADoSj+HZq70EFpa\nqmZ2W1qqljcaIB1reM8ghpGef35yww2r191wgwCJ9TH8GWgqw9YAABi6QQ3PGfawn6Ko/14L/psN\nwIQzbA0AgMYaVGHgTsN+kuTee/szLA4AJpHwCACAoZuZ6W19t1aG/Wzdunr9vn39GRYHAJNIeAQA\nwNANsjDw3FxyyilHr19ergpq99N55/W2HgDaSHgEADDGBjGjWT90Kgz87GcnF11ULW/enFx66fq3\nP6hhcWtdf/3RQdF551Xr+6mp1xGAyaBgNgDAmFqZ0ezIqeunp5s5e9OllyZXXnn0+ksuSa64ovft\nDbtw9iC16ToC0C4KZgMATLj5+dWBQzKYoVud9NpTZmGht/XHM8hhccM2yusIAInwCABgbA1r6NZa\nKz1llpaq6eqXlo5frPqhh7pb320o1WlYXFt76ozqOgLACsPWAADG1KiGbq1nv5s3dw6QNm1KDhyo\nvp7U4VvjNAQPgGYxbA0AYMKNaujWenrK7Nx5/PVNHb416GLW4zQED4B2Eh4BAIypUQ3dmpnpbX1S\nFcW+5JKqp1FSPa4tlt3E4VvrGaLXq3EaggdAOxm2BgBAXw1qeFkTh281sU0A0C3D1gAAhmzQw5fa\nYlA9ZZo4fKuJvaEAoN82j7oBAADjYG1vm5XhS8lkDi+am+v/ca9sb36+CmdmZqrgaJTnd2amc8+j\nYw3RA4C2MWwNAKAPDF+aTJM6AxwA48GwNQCAIRrl8KWmDZdrWnsG6XhD9Pp1LibpnLaR6wOMOz2P\nAAD6YFQ9j5rW86Vp7Rmlfp0L57TZXB+gzbrteSQ8AgDog1HdQDZtuFzT2jNK/ToXzmmzuT5AmwmP\nAACGbHFx+MWcp6aSTv+dK4rk4MHB7rsN7Rmlfp0L57TZXB+gzdQ8AgAYsrm5qqfBwYPVY7+Co2PV\nU6mb1WtUs301rT2j1K9z4Zw2R6fPousDTALhEQD0kaKp42fU13RlONzSUtW7YWmpWl5px65d1fC4\nI01PV+sH1Z5jnY9O7TnhhOTeeyfvc9GvazPsa0xndZ/FF7/Y9QHGn/AIAPrkeDf5tE8Trun8/Oo6\nSkm1PD9ffX282b76qZvzsbY9W7dWz923b/I+F/26NsO8xtSr+yxed53rA4w/NY8AoE8UTR0/Tbim\nTaqnsp7z0YRzCP3QpM8iQL+oeQQAQ7Z3b2/rab4mXNMm1VNZz/lowjmEfmjSZxFg2IRHANAnbizG\nTxOuaZPq3aznfDThHEI/NOmzCDBswiMA6BM3FuOnCde0SfVu1nM+mnAOh2XUxdUZrCZ9FgGGTc0j\nAOijxcWqqOrevVXPil273Fi0nWu62nrOxyScw5Vi4kcWVJ6eFi4A0Gzd1jwSHgEAwAYpDA5AGymY\nDQAAQ6IwOADjTHgEAAAbpDA4AONMeAQAABs0SYXBAZg8wiMAgC4Mcyatun0Nsg2dtm32sO6ZiWs8\neM8DdKZgNgDAcaxnJq31zjBWt6+LLkquvnows3l12ucJJyRlmezf3//9QROZMQ+YRGZbAwDok15n\n0trITWjdvjZtSh56qPs29KJun52YPYxxZcY8YBKZbQ0AoE96nUlrfn51cJRUy/Pz699Xp+DoWM/v\nRS/bmMTZwwxlmgxmzAOoJzwCADiOXmfS2shNaN02N23q7fm96GUbkzZ72EovsqWlahjf0lK1LEAa\nP02ZMU9YCTSR8AgA4Dh6nUlrIzehdfvauXNws3l12ucJJyRbtgxmf22ykV5ktEsTZswTVgJNJTwC\nADiOXmfS2shNaN2+rrhicLN5ddrnVVcl73uf2cMMZZocTZgxT1gJNJWC2QAAA7De2dZoFkWUGaap\nqarH0VpFkRw8OPz2AONPwWwAgBGam6vChYMHq8dugyP1TpqlCUOZmBxNqbsEsJbwCABaSsgwftQ7\naZ4mDGVicggrgaYybA0AWmglZDiyNsb0tJvatjNECjDkFRgmw9YAYIwpqjqe2laceaX3W1EkmzdX\nj932gtNzjnHR7/fyeoe8AgyS8AgAWqhtIQPdaVO9kyOH2CXJQw9Vj90MtZuE4XnCsckwCe9lgMSw\nNQBoJcObxlObhiPWvQdXHOu9OO7v3zZdRzZm3N/LwPgzbA0AxpiiquOpTcWZj9fL7VjfH/eec4aV\nTo5xfy8DrNg86gYAAL1bCRMUVR0/c3PtuI4zM8fueXSsoXZ1r23i8Lz1EChMjnF/L8O4KMsyKZPy\nYPW48u/I5bIsk4NHfF0mOVjzdZJHnP6IUR3OSAiPAKCl2hIyMJ527Tp6aNaKokhe/OJjv/aiiw7X\nSUqSTZvGp+ecQGFydPocDLMX6MrN7sM3tL3eHB/rRnmjN9prn9/HbY30uAZ4jo51zlp1jhp27Qeh\nOKHIcx947mA23lDCIwAAenZk77e1QUlZJldfnZx7bueA87OfXR0cJdXyZz+b/Jt/s+amoEk3W2tf\nX7Otd/x88s53JA8+UKZIUiQ58RFl3vBzybf/sh03W628iRzBtf8XZfLRRyX7Hizz0IFq1sFTT05O\n/rUyf/uWwR4XDVckmUqKosjKD4Ji6vDXa5eLoqiKynT6eqPbOs7rj7mtqeLh166rLRtoV8+vH+A5\nOur1m4vBvG8aTMFsABiCVTcJTbrZ8tfaRh9XW679Xd8qc/Dgyv+pyyTV/8k3FWUe+cijn3//96rn\nHP4/ebny/3aabiM3W2NyEznQG9KmBggbbcsEXfuiKNb76YKR6LZgtp5HwIbU3vBMyk3kuPy1VoAw\n8HNEgx15I9Ckm6313ERuOrxumDdbH3lPDsU/Kx+HQ1+Xyf/xmqPP2e+9Y+WjURzxMale87bfGJ+b\nyKFe+wGfIzfEAJNNz6MhKR8q88CdD7TjJnJM/lrb5JvIcbn2bogbzs3W4G62hnlcY/LX2qEeVy/X\nzg1xX8zO9jZd+ebNRw9bS6q6RwcO9Lt1AEAdPY8aZv8/7c/f/eDfjboZ42HANxJDu9la+UveVNGM\nm61+Hdeob7QFCIfXAwxJr0WDd+5Mrryy83oAoHmER0Oy6ZRNedLCk0ZzEzkuN9puiAGYUIuLVWHq\nvXurGbt27RrOTHvd7vfI4tndtPGKK6rHhYWqB9KmTVVwtLK+bUZ1fQBgWAxbAwBosMXFzr16FhYG\nG1CMar9t4zwB0GbdDlszqQUAQIPNz68OJpJqeX5+PPfbq8XFZHY2mZqqHhcXh7v/tpwnANgIw9YA\nABps797e1rd9v71Y2+tnaelw3aRh9fppw3kCgI3S8wgAoMFmZnpb3/b99qIJvX7acJ4AYKOER9AA\nl15aTVtcFNXjpZeOukUArFe/h1Ht2lXV0DnSkTOZDWLY1uJicu+9R68viqp3zyiGh3XShF4/u3ZV\nv7uPtHlz/Uxzk2DUQwnHjfMJNIHwCEbs0kur6YofeqhafuihalmABNA+K8OolpaSsjw8jGojN3tz\nc1Xx5W3bqvBm27bDxZgHsb+Vbe7bd/T3VuZZ6cd++qEJvX4++9nkwIHV6w4cqNZPokG8JyeZ8wk0\nhdnWYMQ2bz4cHB1p06aj/zMKQLPNzlY3d2tt25bs2dOO/dVts5NBHVe3mjDTmd/jqw37MzDunE9g\n0My2Bi3R6T+cx1oPQHMNexjVIPbXy2tHXRT6WL2yhuVYv8cncahRE4YSjhPnE2gK4RGM2KZNva0H\noLmGPYxqEPvr5bVNKAo9N1f1wDh4sHocZnCUHPv39SQONerlPamWz/E1YWgmQCI8gpFbmVK42/UA\nNNfxilu3YX+dtrllS3LCCf3dz7jo9vf1sGeBG5Vu35Nq+XRn2D9TAOoIj2DErrgiueSSw3+53LSp\nWr7iitG2C4BKL70jBjWMqq4N3e5vo8fwvvclV13V3+Mal14nnX6P15mEoUbdvifn51fXqkomJ2Dr\nRROGZgIkCmYDANRqQkHmjbahCcfQhjZt1OJiFXzs3VsFYp1qISlyfNjU1OHZ+45UFNUQRACGo9uC\n2cIjAIAaTZjpaKNtaMIxrNXENm1EpzBsrbaHY/02bu8BgLYy2xoAwAY1YaajjbahCcfQ7b7bOqyr\n0xCspBrCZqhRZ2r5ALSL8AgAoEYTZjraaBuacAzd7rutM0jVhV4HD45uFrimU8sHoF2ERzBA41IM\nFGBSNaF3xEbbsJHXD+r3WK9tavrv03ELw4Zlbq4K1gRsAM0nPIIBMQUtQPs1oXfERtuw3tcP8vdY\nL21qw+/TJoSMADBICmbDgCgECUCbNeX3WFPacTxHzrY2M1MFR3rSANB0ZluDETMFLQBt1pTfY01p\nBwCMI7OtwYipfwBAmzXl91hT2gEAk2xg4VFRFD9YFMX/WxTFrUVR/GNRFL9yaP2pRVF8siiK2w89\nPmZQbYBRUv8AYLI0vahzrwb9e+zSS5PNm6seRJs2Jaec0vncdWpHktx7b3PO8bhdewBYa5A9jw4k\n+XdlWf5IkmcleV1RFD+S5LIkN5Rl+cQkNxxahrHThCKrwHhyo9o8bSjq3KtB/h679NLkyiuThx6q\nlg8eTO67r/O5W2nH1q2rt7FvXzPO8TheewBYa2g1j4qi+GiSdx3697yyLO8siuL0JJ8qy/LMY71W\nzSMAqKzcqC4vH143PS2cHrW2FHUelF6LRW/efDg4qrP23PXjHA+iqPWkX3sA2q1RBbOLophN8ukk\nT0mytyzL7zu0vkjyTyvLa16zM8nOJJmZmXnmUqffygAwYdyoNtMkF3VeT6BZFMff7tpzt9FzPKjg\ndZKvPQDt15iC2UVRnJLkT5P8almW3znye2WVXHVMr8qyXCjLckdZljtOO+20QTcTAFph797e1jMc\nk1zUeX5+dSCTVMvz8/Wv2bTp+Ntde+42eo7X085uTPK1B2ByDDQ8KopiS6rgaLEsyz87tPqbh4ar\n5dDjtwbZBgAYJ25Um2mSJ0lYT6C5c+fxt/viF69e3ug5HlTwOsnXHoDJMcjZ1ook703ypbIsf/+I\nb30syUWHvr4oyUcH1QYAGDduVJupLZMkDKLY+noCzSuuSM4779jbve661csbPceDCl7bcu0BYCMG\nVvOoKIofS/KZJF9IsjLi+9eS/H2SDyeZSbKU5JVlWX77WNtSMBsADhtE0V/G36Bq/qxnu51es1a/\nawYpNg8AR2tUweyNEh4BAGzMIIut9xpo1rWl3+1aS/AKAKsJjwAAeFiTZgWra8sKPYIAYDgaM9sa\nAACj16Ri66eeWv89NYMAoHmERwAAE6ANxda3bq2Gqs3NDaa497hwbgAYts2jbgAAAIO30pOnCTV/\nvl0zVcq+fYfrIRXF4aFtS0tVsetEj6S1hb+dGwCGQc0jAACGqq5g9pGBUSeDKKLdNoMsfA7A5FHz\nCABgArRxCFOnIXTHC46SqsfUpKs7B84NAIMkPAIAaKmVIUxLS1XwsjKEqekB0txcVRR727YqNNq2\n7fjBUTKa4t5N06TC5wBMDuERAEBLzc8frn2zYnm5Wt90c3PVMKuDB6ueSJs2Hfv5TSvuPSptKHwO\nwPgRHgEAtFTdUKWlpfYMY1vpPfXQQ8d+3kknDac9/TKo4YSdem0tLCiWDcBgKZgNANBSdcWTjzQ9\n3exwoe4Ypqaq3kj79x9e1/RjWbF2RrSkPW0HYLJ0WzBbeAQA0FKdQopOmjwT19RUd/WOVjT5WFaY\nEQ2AtjDbGgDAmFs7hKlOk2fi6rXQc5OPZYUZ0QAYN8IjAIAWO7Lw9LZtnZ/T5Jm4OhWA3rKl/vlN\nPpYVZkQDYNwIjwAAxkSbZuK69NJk8+bkwguT++9PTj75cAHoRz2q82uKopnHsrY49otf3J7rAADd\nEB4BAIyJtszEdemlyZVXHp5h7eDB5L77kl/+5aoX1be/3fl1Zdm8Y1mpO7W0VLVvaSm5+urkooua\nfx0AoFsKZgMAMFSbNx8Ojo60aVNy4EC7Ck63qa0AsJaC2QAANFKn4OjI9bt2HV33aMuWZg77Uhwb\ngEkgPAIAYKg2bTr++rWzxx1rNrlRUhwbgEkgPAIA2KC1BZMXF0fdombbufPY6+fnkwcfXP29Bx+s\n1jdNm4qUA8B6CY8AADagU8HknTsFSMdyxRXJJZcc7mm0aVO1fMUV1XKbhoK1pUg5AGyEgtkAABsw\nyQWTFxer3kB791bDtHbt6k9oMsnnFACGScFsAIAhaFMvmX4aZI8rQ8EAoFmERwAAGzCpBZPn55Pl\n5dXrlpf7U5fIUDAAaBbhEQDABkxqL5lOw8qS/vW4mpurhqgdPFg9Co4AYHSERwAAGzCJvWQWF6tj\n7WRqavDFws1uBwDDpWA2AAA9qStovWJ6enAB2kqtpSOHzA1yfwAwzrotmC08AgCgJ1NTVZHsYxnU\nzGhmYgOA/jHbGgAAA9FNMfBBzTY3qbPbAcAoCY8AAOhJpyLhaw1qtrlJnd0OAEZJeAQAQE+OLBKe\nHF08e5CzzU3q7HYAMErCIwCAdZrkWb/m5qoaQ2WZfOADw5ttbhJntwOAUVMwGwBgHcz61dniYjI/\nX9UgmpmpegRN8vkAgCZTMBsAYIDm51cHR0m1PD8/mvY0wUqgtrRU9UhaWqqWJ6lHFgCMI+ERAMA6\nmPXraAI1ABhPwiMAgHUw69fRBGoAMJ6ERwAA62DWr6MJ1ABgPAmPAADWwaxfRxOoAcB42jzqBgAA\ntNXc3GSHRWutnAuzrQHAeBEeAQDQNwI1/v/27j7GsrusA/j32Rai05JYaNNUSneBkBBUxLhiwpuA\nqGDUliiEOoloDCsJGjTBSFxfiGETikqMCb6sEXlxlAgINEBUUF7qC9BdbKGFIE3pLjSVtgGEusZK\n+/jHPZvOLnt3drY798zc8/kkk3vPb+8989s8+c3s/e7vPAeA5eOyNQAAAADmEh4BAAAAMJfwCACA\nUaytJXv2JLt2zR7X1saeEeeS+gIsDz2PAABYuLW1ZN++5Nix2fGRI7PjRM+kZaC+AMulunvsOWxo\n7969fejQobGnAQDAObJnzyxQONnu3cltty16Npxr6guwM1TV4e7eu9HrXLYGAMDCHT26uXF2FvUF\nWC7CIwAAFu6KKzY3zs6ivgDLRXgEAMDCHTiQrKycOLayMhvn1HZSA2r1BVguwiMAtsxO+qADLNbq\nanLw4KwHTtXs8eBBzZTnOd6A+siRpPuBBtTb9eeq+gIsFw2zAdgSJ99pJ5n9r7MPDwCbpwE1AFtB\nw2wARrV//4nBUTI73r9/nPkA24udiZujATUAYxIeAbAlfNAB5tlpl2BtBxpQAzAm4REAW8IHHWAe\nOxM3TwNqAMYkPAJgS/igA8xjZ+LmaUANwJiERwBsCR90gHnsTDw7q6uz5tj33z979PMUgEURHgGw\nZXzQAU7FzkQA2FmERwDsWO7WBDuTnYkAsLMIjwDYkdytCc7edghe7UwEgJ1DeATAjuRuTXB2tnPw\nuh1CLQDgm1V3jz2HDe3du7cPHTo09jQA2EZ27Zp98D1Z1WwnA3Bqe/bMAqOT7d492wE0luOh1vpQ\neGXF5WwAsJWq6nB3793odXYeAbAjuVsTnJ2jRzc3vih2EwLA9iU8AmBHcrcmODvbNXjdrqEWACA8\nAmAbOJs+J+7WBGdnuwav2zXUAgCERwCM7ME073W3Jti87Rq8btdQCwDQMBuAkW3X5r3A4q2tzXoc\nHeRE3mgAAArQSURBVD0623F04MD4oRYALLMzbZgtPAJgVO6aBgAA43C3NQB2BH1OAABgexMeATCK\n402yjxyZ7TJaT58TAADYPoRHACzc+ibZyeyyteMB0nZp3gsAAMycP/YEAJie/fuTY8dOHOvWJBsA\nALYjO48AWLijRzc3DgAAjEd4BMDCaZINAAA7h/AIgIU7cGDWFHs9TbIBAGB7Eh4BsHCrq7Om2Lt3\nzxpla5INAADbl4bZAIxidVVYBAAAO4GdRwAAAADMJTwCYHLW1pI9e5Jdu2aPa2tjzwgAALYvl60B\nMClra8m+fcmxY7PjI0dmx4nL6AAA4FTsPAJgUvbvfyA4Ou7Ysdk4AADwzYRHAEzK0aObGwcAgKkT\nHgEwKVdcsblxAACYOuERAJNy4ECysnLi2MrKbBwAAPhmwiMAJmV1NTl4MNm9O6maPR48qFk2AADM\n425rAEzO6qqwCAAAzpSdRwAslbW1ZM+eZNeu2ePa2tgzAgCAnc3OIwCWxtpasm9fcuzY7PjIkdlx\nYqcRAACcLTuPAFga+/c/EBwdd+zYbBwAADg7wiMAlsbRo5sbBwAANiY8AmBpXHHF5sYBAICNCY8A\nWBoHDiQrKyeOrazMxgEAgLMjPAJgaayuJgcPJrt3J1Wzx4MHNcsGAIAHw93WAFgqq6vCIgAAOJfs\nPAIAAABgLuERAAAAAHMJjwAAAACYS3gEAAAAwFzCIwAAAADmEh4BAAAAMJfwCAAAAIC5hEcAAAAA\nzCU8AgAAAGCuUcKjqnpuVX22qm6pqleOMQcAAAAANrbw8Kiqzkvy+iTPS/KEJFdX1RMWPQ8AAAAA\nNjbGzqMnJ7mlu2/t7nuTvDXJlSPMAwAAAIANjBEePTLJF9Ydf3EYO0FV7auqQ1V16K677lrY5AAA\nAAB4wLZtmN3dB7t7b3fvveSSS8aeDgAAAMAkjREe3Z7kUeuOLx/GAAAAANhmxgiPrk/yuKp6dFU9\nNMmLklw7wjwAAAAA2MD5i/6G3f2NqvrFJH+f5Lwkb+jumxc9DwAAAAA2tvDwKEm6+31J3jfG9wYA\nAADgzG3bhtkAAAAAjK+6e+w5bKiq7kpyZOx5nCMXJ7l77EmwcOo+XWo/XWo/XWo/XWo/XWo/Teo+\nXctU+93dveEt7ndEeLRMqupQd+8dex4slrpPl9pPl9pPl9pPl9pPl9pPk7pP1xRr77I1AAAAAOYS\nHgEAAAAwl/Bo8Q6OPQFGoe7TpfbTpfbTpfbTpfbTpfbTpO7TNbna63kEAAAAwFx2HgEAAAAwl/AI\nAAAAgLmERwtWVa+qqtur6obh60fHnhNbq6qeW1WfrapbquqVY8+Hxamq26rqU8NaPzT2fNg6VfWG\nqrqzqm5aN/bwqnp/VX1ueLxozDmyNebU3u/6JVdVj6qqD1bVp6vq5qp6+TBu3S+509Teul9yVfUt\nVfXxqrqxqj5TVa8Zxq37JXea2k9q3et5tGBV9aok93T37409F7ZeVZ2X5D+S/FCSLya5PsnV3f3p\nUSfGQlTVbUn2dvfdY8+FrVVVz0hyT5I3d/d3DmOvTfLl7n7NEBxf1N2/NuY8Offm1P5V8bt+qVXV\nZUku6+5PVNXDkhxOclWSn411v9ROU/sXxrpfalVVSS7o7nuq6iFJ/jnJK5L8eKz7pXaa2v9gJrTu\n7TyCrfXkJLd0963dfW+Stya5cuQ5AedYd38kyZdPGr4yyZuG52/K7MMFS2ZO7Vly3X1Hd39ieP71\nJJ9J8shY90vvNLVnyfXMPcPhQ5Kcl+Qrse6X3mlqPynCo3H8UlV9ctjqblvjcntkki+sO/5i/ANj\nSjrJB6rqcFXtG3syLNyl3X3H8Pw/k1w65mRYOL/rJ6Kq9iT5niQfi3U/KSfVPrHul15VnVdVNyS5\nM8mHuvumWPeTMKf2yYTWvfBoC1TVB6rqplN8XZnkj5M8JsmTktyR5PdHnSywlZ7W3U9K8rwkLxsu\nb2GCenaNuOvEp8Pv+omoqguTvCPJL3f319b/mXW/3E5Re+t+Arr7vuHfdpcneXpVPeukP7ful9Sc\n2k9q3Z8/9gSWUXc/50xeV1V/luQ9WzwdxnV7kketO758GGMCuvv24fHOqnpnZpcxfmTcWbFAX6qq\ny7r7jqFHxp1jT4jF6O4vHX/ud/3yGvpevCPJWnf/7TBs3U/AqWpv3U9Ld3+1qt6bZG+s+0lZX/vu\n/uDx8SmsezuPFmz4gXLc85PcNO+1LIXrkzyuqh5dVQ9N8qIk1448Jxagqi4YGmmmqi5I8sOx3qfm\n2iQvHp6/OMm7R5wLC+R3/fIbmqf+eZLPdPfr1v2Rdb/k5tXeul9+VXVJVX3b8PxbM7shzg2x7pfe\nvNpPbd2729qCVdVbMtvW1kluS/IL666RZQkNt2z8g8waq72huw+MPCUWoKoek+Sdw+H5Sf5K7ZdX\nVf11kmcmuTjJl5L8dpJ3JfmbJFckOZLkhd2tsfKSmVP7Z8bv+qVWVU9Lcl2STyW5fxj+9cx631j3\nS+w0tb861v1Sq6onZtYQe9fw9ZfdfU1VPSLW/VI7Te0n9dleeAQAAADAXC5bAwAAAGAu4REAAAAA\ncwmPAAAAAJhLeAQAAADAXMIjAAAAAOYSHgEAnANVdVVVPeFBvH9PVf30g5zDgz4HAMDJhEcAwCRU\n1flb/C2uSnLW4VGSPUkebPBzLs4BAHAC4REAsDSq6meq6pNVdWNVvaWq3lhVf1JVH0vy2jnvubCq\n/qKqPjW89yeH8auHsZuq6pp1r7+nqg4M3+OjVXVpVT0lyU8k+d2quqGqHjt8/V1VHa6q66rq8cP7\n31hVf1hV/1pVt1bVTw2nfk2Spw/v/5U5c31vVT1xeP7vVfVbw/PfqaqXnMk5AAA2a6v/Bw4AYCGq\n6juS/EaSp3T33VX18CSvS3L5MHbfnLf+ZpL/6u7vGs5zUVV9e5Jrknxvkq8k+Yequqq735XkgiQf\n7e79VfXaJC/p7ldX1bVJ3tPdbx/O849JXtrdn6uq70/yR0mePXzPy5I8Lcnjk1yb5O1JXpnkFd39\nY6f5a16XWTh0JMk3kjx1GH96kpcm+dwZnAMAYFPsPAIAlsWzk7ytu+9Oku7+8jD+ttMER0nynCSv\nP37Q3V9J8n1JPtTdd3X3N5KsJXnG8JJ7k7xneH44s0vFTlBVFyZ5SpK3VdUNSf40s8DouHd19/3d\n/ekkl27i73jdMI+nJnlvkguraiXJo7v7s5s4DwDAGbPzCABYdv99js/3f93dw/P7cup/T+1K8tXu\nftKcc/zvuue1ie99fZK9SW5N8v4kFyd5SWYhFgDAlrDzCABYFv+U5AVV9YgkGS5bOxPvT/Ky4wdV\ndVGSjyf5gaq6uKrOS3J1kg9vcJ6vJ3lYknT315J8vqpeMJyzquq7z/T983T3vUm+kOQFSf4ts51I\nr0jykTM9BwDAZgmPAICl0N03JzmQ5MNVdWNm/Y7OxKuTXDQ0xr4xybO6+47MehB9MMmNSQ5397s3\nOM9bk/zq0Mj6sUlWk/z8cM6bk1y5wfs/meS+oRH36ZpdX5fkzu7+n+H55cPjZs4BAHDG6oFd1wAA\nAABwIjuPAAAAAJhLw2wAYBKq6ueSvPyk4X/p7ped6vVjqqofSXLNScOf7+7njzEfAGDaXLYGAAAA\nwFwuWwMAAABgLuERAAAAAHMJjwAAAACYS3gEAAAAwFzCIwAAAADm+n83rnoQg8B4OwAAAABJRU5E\nrkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2d568a86e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r_square value for cr= 0.0116706237977\n",
      "reverified value for r square for cr= 0.0393106285513\n"
     ]
    }
   ],
   "source": [
    "##########################################################################\n",
    "def lin_reg3():\n",
    "    print(\"we will check how other components in the alloy and SFE are corelated\")\n",
    "    #####################\n",
    "    #rehashing same code from above\n",
    "    ####################\n",
    "    plt.rcParams['figure.figsize']=(20.0,10.0)\n",
    "    data = pd.read_csv('sfe_ni_cr.csv')\n",
    "    #to have a heads up about the data we are dealing with\n",
    "\n",
    "    print (\"######################################################################################## \\nPerforming Linear regression on dataset 2\")\n",
    "    print (\"More about the dataset:\\nFrom paper 'Composition-dependence of stacking fault energy in austenitic stainless steels through linear regression with random intercepts' \")\n",
    "    print(data.shape)\n",
    "    #look at the first five rows of data for easier programming\n",
    "    data.head()\n",
    "\n",
    "    #appending values from columns into arrays\n",
    "    cr_content_wt=data['Cr'].values\n",
    "    SFE=data['SFE'].values\n",
    "    #print(cr_content_wt) \n",
    "    #print(SFE)\n",
    "\n",
    "    xarr2a=np.array(cr_content_wt)\n",
    "    yarr2a=np.array(SFE)\n",
    "\n",
    "    #calculating the regression line\n",
    "    #intercept and slope of the line \n",
    "    cr_cont_mean=np.mean(cr_content_wt)\n",
    "    SFE_mean=np.mean(SFE)\n",
    "    mm2=len(cr_content_wt)\n",
    "    jj2=0\n",
    "    kk2=0\n",
    "    for i in range(mm2):\n",
    "        jj2+=(cr_content_wt[i]-cr_cont_mean)*(SFE[i]-SFE_mean)\n",
    "        kk2+=(cr_content_wt[i]-cr_cont_mean)**2\n",
    "\n",
    "    #slope of the line\n",
    "    slope_2a=jj2/kk2 \n",
    "\n",
    "    #intercept\n",
    "    intercept_2a=SFE_mean - (slope_2a*ni_cont_mean) \n",
    "\n",
    "    print (slope_2a,intercept_2a)\n",
    "\n",
    "    #plotting the values and the fit line together\n",
    "    #for the plot\n",
    "    xlim2a=np.max(cr_content_wt)+5\n",
    "    ylim2a=np.min(cr_content_wt)-5\n",
    "    x2a=np.linspace(ylim2a,xlim2a,1000)\n",
    "\n",
    "    #equation of line\n",
    "    y2a=intercept_2a+slope_2a*x2\n",
    "\n",
    "    print(\"equation of best  fit line=\",intercept_2a,\"+\",slope_2a,\"x\")\n",
    "    plt.plot(x2a,y2a,color='m', label='regression line')\n",
    "    plt.scatter(xarr2a,yarr2a,c='b',label='scatter plot')\n",
    "\n",
    "    plt.xlabel('cr_content_wt')\n",
    "    plt.ylabel('SFE')\n",
    "    plt.legend()\n",
    "    plt.show()\n",
    "\n",
    "    ############################TESTING###################################\n",
    "    #r square method to determine the fit of the line to the data\n",
    "    sum_sq_total2a =0\n",
    "    sum_sq_res2a =0\n",
    "\n",
    "    for i in range(mm2):\n",
    "        pred_eq2a = intercept_2a+slope_2a*xarr2a[i]\n",
    "        sum_sq_total2a+=(yarr2a[i]-SFE_mean)**2\n",
    "        sum_sq_res2a+=(yarr2a[i]-pred_eq2a)**2\n",
    "\n",
    "    r_square2a=1-(sum_sq_res2a/sum_sq_total2a)\n",
    "\n",
    "    print(\"r_square value for cr=\",r_square2a)\n",
    "\n",
    "    #verification trough scipy to check the value of r is correct\n",
    "\n",
    "    slope, intercept, r_value, p_value, std_err = stats.linregress(xarr2a, yarr2a)\n",
    "    re_verif_value2a = r_value ** 2\n",
    "    print (\"reverified value for r square for cr=\",re_verif_value2a)\n",
    "    np.isclose(r_square2a, re_verif_value2a, rtol=1e-13, atol=0.0)\n",
    "lin_reg3()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Linear regression to predict the pricing of houses based on dataset where it is assumed to be depedendent on \n",
      "one variable only(lotsizes) for the simplicity of the problem\n"
     ]
    }
   ],
   "source": [
    "###########################################################################################################\n",
    "print(\"Linear regression to predict the pricing of houses based on dataset where it is assumed to be depedendent on \\none variable only(lotsizes) for the simplicity of the problem\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(546, 13)\n",
      "0.0435046391184 2186.66007109\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKIAAAJQCAYAAAC97syiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAIBJKEQBAAAAMIn/AXxRvjzfm95BAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2d568a99940>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.287077002597\n",
      "reverified value for r sq= 0.287077002597\n"
     ]
    }
   ],
   "source": [
    "def lin_reg_housing():\n",
    "    plt.rcParams['figure.figsize']=(20.0,10.0)\n",
    "    data = pd.read_csv('housing.csv')\n",
    "    print(data.shape)\n",
    "    data.head()\n",
    "    x=data['price'].values\n",
    "    y=data['lotsize'].values\n",
    "    xarr=np.array(x)\n",
    "    yarr=np.array(y)\n",
    "    mean_x=np.mean(x)\n",
    "    mean_y=np.mean(y)\n",
    "    m=len(x)\n",
    "    numer=0\n",
    "    denom=0\n",
    "    for i in range(m):\n",
    "        numer+=(x[i]-mean_x)*(y[i]-mean_y)\n",
    "        denom+=(x[i]-mean_x)**2\n",
    "    b1=numer/denom\n",
    "    b0=mean_y - (b1*mean_x)\n",
    "    print (b1,b0)\n",
    "    max_x=np.max(x)+100\n",
    "    min_x=np.min(x)-100\n",
    "    x=np.linspace(min_x,max_x,1000)\n",
    "    y=b0+b1*x\n",
    "    plt.plot(x,y,color='r', label='regression line')\n",
    "    plt.scatter(xarr,yarr,c='g',label='scatter plot')\n",
    "    plt.xlabel('price')\n",
    "    plt.ylabel('lotsize')\n",
    "    plt.legend()\n",
    "    plt.show()\n",
    "    #rms method\n",
    "    ss_t =0\n",
    "    ss_r =0\n",
    "    for i in range(m):\n",
    "        y_pred = b0+b1*xarr[i]\n",
    "        ss_t+=(yarr[i]-mean_y)**2\n",
    "        ss_r+=(yarr[i]-y_pred)**2\n",
    "    r2=1-(ss_r/ss_t)\n",
    "    print(r2)\n",
    "    slope, intercept, r_value, p_value, std_err = stats.linregress(xarr, yarr)\n",
    "    re_verif_value = r_value ** 2\n",
    "    print (\"reverified value for r sq=\",re_verif_value)\n",
    "    np.isclose(r2, re_verif_value, rtol=1e-13, atol=0.0)\n",
    "lin_reg_housing()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10, 7)\n",
      "[ 38.5  20.   19.4  13.7  11.    6.8   6.7   6.3   5.4   5. ]\n",
      "3505.11580911 -5463.03794493\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKkAAAJQCAYAAABFDjdVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nn1geR8TnAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2d5689f50b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.839790991318\n",
      "reverified value for r sq= 0.839790991318\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def lin_reg_death():\n",
    "    plt.rcParams['figure.figsize']=(20.0,10.0)\n",
    "    data = pd.read_csv('USCS_TopTen_deaths(lin).csv')\n",
    "    print(data.shape)\n",
    "    data.head()\n",
    "    x=data['AgeAdjustedRate'].values\n",
    "    y=data['CaseCount'].values\n",
    "    print (x)\n",
    "    xarr=np.array(x)\n",
    "    yarr=np.array(y)\n",
    "    mean_x=np.mean(x)\n",
    "    mean_y=np.mean(y)\n",
    "    m=len(x)\n",
    "    numer=0\n",
    "    denom=0\n",
    "    for i in range(m):\n",
    "        numer+=(x[i]-mean_x)*(y[i]-mean_y)\n",
    "        denom+=(x[i]-mean_x)**2\n",
    "    b1=numer/denom\n",
    "    b0=mean_y - (b1*mean_x)\n",
    "    print (b1,b0)\n",
    "    max_x=np.max(x)+1\n",
    "    min_x=np.min(x)-1\n",
    "    x=np.linspace(min_x,max_x,1000)\n",
    "    y=b0+b1*x\n",
    "    plt.plot(x,y,color='r', label='regression line')\n",
    "    plt.scatter(xarr,yarr,c='g',label='scatter plot')\n",
    "    plt.xlabel('Age Adjusted rate')\n",
    "    plt.ylabel('CaseCount')\n",
    "    plt.legend()\n",
    "    plt.show()\n",
    "    #rms method\n",
    "    ss_t =0\n",
    "    ss_r =0\n",
    "    for i in range(m):\n",
    "        y_pred = b0+b1*xarr[i]\n",
    "        ss_t+=(yarr[i]-mean_y)**2\n",
    "        ss_r+=(yarr[i]-y_pred)**2\n",
    "    r2=1-(ss_r/ss_t)\n",
    "    print(r2)\n",
    "    slope, intercept, r_value, p_value, std_err = stats.linregress(xarr, yarr)\n",
    "    re_verif_value = r_value ** 2\n",
    "    print (\"reverified value for r sq=\",re_verif_value)\n",
    "    if np.isclose(r2, re_verif_value, rtol=1e-13, atol=0.0):\n",
    "        return True\n",
    "lin_reg_death()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "multivariate linear regression\n",
      "(546, 13)\n",
      "(546, 3)\n",
      "[[ 0.  0.  0.]]\n",
      "(546, 1)\n",
      "(546, 3)\n",
      "[[ -2.22923023e-16   4.91909822e-01   2.92099074e-01]]\n",
      "0.314865328335\n",
      "(546, 13)\n"
     ]
    }
   ],
   "source": [
    "#######################################################################################\n",
    "def mult_reg():\n",
    "    print(\"multivariate linear regression\")\n",
    "\n",
    "    ###########################################################\n",
    "    #######adjusting the data##################################\n",
    "    #multivariate linear regression\n",
    "    my_data = pd.read_csv('Housing.csv') #read the data\n",
    "    print (my_data.shape)\n",
    "    my_data.head()\n",
    "\n",
    "    x=my_data\n",
    "    #print(x)\n",
    "\n",
    "    newdf = x[x.columns[1:4]]\n",
    "    #print(newdf)\n",
    "\n",
    "    newdf = (newdf - newdf.mean())/newdf.std()\n",
    "    print (newdf.shape)\n",
    "\n",
    "    newdf.head()\n",
    "\n",
    "    #setting the matrixes\n",
    "\n",
    "    y = newdf.iloc[:,1:3]\n",
    "    theta = np.zeros([1,3])\n",
    "\n",
    "    print (theta)\n",
    "    X = my_data.iloc[:,1:2].values #.values converts it from pandas.core.frame.DataFrame to numpy.ndarray\n",
    "\n",
    "    ones = np.ones([y.shape[0],1])\n",
    "    y = np.concatenate((ones,y),axis=1)\n",
    "\n",
    "\n",
    "    #set hyper parameters\n",
    "\n",
    "    alpha = 0.01\n",
    "\n",
    "    iters = 1000\n",
    "\n",
    "    X = (X - X.mean())/X.std()\n",
    "\n",
    "    #print(X)\n",
    "    print (X.shape)\n",
    "\n",
    "    #print (y)\n",
    "    print (y.shape)\n",
    "\n",
    "    def computeCost(X,y,theta):\n",
    "\n",
    "        tobesummed = np.power(((y @ theta.T)-X),2)\n",
    "\n",
    "        return np.sum(tobesummed)/(2 * len(y))\n",
    "    def gradientDescent(X,y,theta,iters,alpha):\n",
    "\n",
    "        cost = np.zeros(iters)\n",
    "\n",
    "        for i in range(iters):\n",
    "\n",
    "            theta = theta - (alpha/len(y)) * np.sum(y * (y @ theta.T - X), axis=0)\n",
    "\n",
    "            cost[i] = computeCost(X,y,theta)\n",
    "\n",
    "\n",
    "\n",
    "        return theta,cost\n",
    "\n",
    "\n",
    "\n",
    "    #set hyper parameters\n",
    "\n",
    "    alpha = 0.01\n",
    "\n",
    "    iters = 1000\n",
    "\n",
    "\n",
    "\n",
    "    g,cost = gradientDescent(X,y,theta,iters,alpha)\n",
    "\n",
    "    print(g)\n",
    "\n",
    "\n",
    "\n",
    "    finalCost = computeCost(X,y,g)\n",
    "\n",
    "    print(finalCost)\n",
    "\n",
    "    %matplotlib inline\n",
    "    my_data = pd.read_csv('Housing.csv') #read the data\n",
    "    print (my_data.shape)\n",
    "    #my_data.head()\n",
    "    x=my_data\n",
    "    #print(x)\n",
    "\n",
    "\n",
    "    newdf = x[x.columns[1:4]]\n",
    "    #print(newdf)\n",
    "    lotsize=np.array(newdf.iloc[:,1])\n",
    "    #print (lotsize)\n",
    "    price = np.array(newdf.iloc[:,0])\n",
    "    #print (price)\n",
    "    bedrooms = np.array(newdf.iloc[:,1])\n",
    "    #print (bedrooms)\n",
    "\n",
    "    \n",
    "            \n",
    "mult_reg()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Pawan\\AppData\\Local\\Enthought\\Canopy\\edm\\envs\\User\\lib\\site-packages\\statsmodels\\compat\\pandas.py:56: FutureWarning: The pandas.core.datetools module is deprecated and will be removed in a future version. Please use the pandas.tseries module instead.\n",
      "  from pandas.core import datetools\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(546, 13)\n",
      "        price   lotsize  bedrooms\n",
      "0   -0.978239  0.322732  0.047192\n",
      "1   -1.109312 -0.530526 -1.308947\n",
      "2   -0.697368 -0.964074  0.047192\n",
      "3   -0.285425  0.691709  0.047192\n",
      "4   -0.266700  0.557955 -1.308947\n",
      "5   -0.079453 -0.456731  0.047192\n",
      "6   -0.079453 -0.585873  0.047192\n",
      "7    0.032896 -0.456731  0.047192\n",
      "8    0.587147 -0.161550  0.047192\n",
      "9    0.763160  0.161305  0.047192\n",
      "10   0.819334  0.945380  0.047192\n",
      "11  -1.408908 -0.991747 -1.308947\n",
      "12  -1.539981 -1.591334  0.047192\n",
      "13  -1.202936 -1.047094  0.047192\n",
      "14  -1.165486 -0.715015 -1.308947\n",
      "15  -1.131782 -0.906421 -1.308947\n",
      "16  -1.034413 -0.853381  0.047192\n",
      "17  -1.025051  0.022939  1.403330\n",
      "18  -0.865891 -0.784198 -2.665085\n",
      "19  -0.865891 -0.536984 -1.308947\n",
      "20  -0.734818 -0.168468  0.047192\n",
      "21  -0.083198 -0.295304  1.403330\n",
      "22  -1.131782 -0.530526  0.047192\n",
      "23  -1.128037 -0.560967 -1.308947\n",
      "24  -0.978239 -0.087754 -1.308947\n",
      "25  -0.967004 -0.991747 -1.308947\n",
      "26  -0.922065 -0.622771 -1.308947\n",
      "27  -0.903340 -0.087754 -1.308947\n",
      "28  -0.884616 -0.991747  0.047192\n",
      "29  -0.869636 -0.299916  0.047192\n",
      "..        ...       ...       ...\n",
      "516  1.006581  0.391915  0.047192\n",
      "517  1.062755  0.391915  1.403330\n",
      "518  1.231278  0.502608  1.403330\n",
      "519  1.268727  0.391915  0.047192\n",
      "520  1.306177  1.166766  1.403330\n",
      "521  1.381075  0.391915  1.403330\n",
      "522  1.493424  0.391915  1.403330\n",
      "523  1.568323  0.391915  1.403330\n",
      "524  1.680671  0.391915  1.403330\n",
      "525  1.942817  1.072216  0.047192\n",
      "526  1.381075 -0.000122  0.047192\n",
      "527  1.418525  0.541812  0.047192\n",
      "528  1.474699  0.391915  0.047192\n",
      "529  1.493424  0.391915  0.047192\n",
      "530  1.708758  0.391915  0.047192\n",
      "531  1.942817  0.853136  0.047192\n",
      "532  0.070345  3.574339  0.047192\n",
      "533  0.107795  1.169534  0.047192\n",
      "534  0.519738 -0.069306  0.047192\n",
      "535  0.519738  0.299671  0.047192\n",
      "536  0.538463  0.391915  0.047192\n",
      "537  0.557188 -0.161550  0.047192\n",
      "538  0.594637  0.622526  0.047192\n",
      "539  0.632087  1.000727  1.403330\n",
      "540  0.632087  0.634056  0.047192\n",
      "541  0.875508 -0.161550  0.047192\n",
      "542  0.969132  0.391915  0.047192\n",
      "543  1.306177  0.391915  0.047192\n",
      "544  1.381075  0.391915  0.047192\n",
      "545  1.381075  0.391915  0.047192\n",
      "\n",
      "[546 rows x 3 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Pawan\\AppData\\Local\\Enthought\\Canopy\\edm\\envs\\User\\lib\\site-packages\\ipykernel\\__main__.py:14: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "C:\\Users\\Pawan\\AppData\\Local\\Enthought\\Canopy\\edm\\envs\\User\\lib\\site-packages\\pandas\\core\\frame.py:2356: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self.loc._setitem_with_indexer((slice(None), indexer), value)\n",
      "C:\\Users\\Pawan\\AppData\\Local\\Enthought\\Canopy\\edm\\envs\\User\\lib\\site-packages\\pandas\\core\\frame.py:2326: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self._setitem_array(key, value)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      lotsize  bedrooms\n",
      "0    0.323028  0.047235\n",
      "1   -0.531013 -1.310147\n",
      "2   -0.964958  0.047235\n",
      "3    0.692343  0.047235\n",
      "4    0.558466 -1.310147\n",
      "5   -0.457150  0.047235\n",
      "6   -0.586410  0.047235\n",
      "7   -0.457150  0.047235\n",
      "8   -0.161698  0.047235\n",
      "9    0.161453  0.047235\n",
      "10   0.946247  0.047235\n",
      "11  -0.992657 -1.310147\n",
      "12  -1.592794  0.047235\n",
      "13  -1.048054  0.047235\n",
      "14  -0.715670 -1.310147\n",
      "15  -0.907253 -1.310147\n",
      "16  -0.854164  0.047235\n",
      "17   0.022960  1.404617\n",
      "18  -0.784917 -2.667529\n",
      "19  -0.537476 -1.310147\n",
      "20  -0.168623  0.047235\n",
      "21  -0.295575  1.404617\n",
      "22  -0.531013  0.047235\n",
      "23  -0.561481 -1.310147\n",
      "24  -0.087835 -1.310147\n",
      "25  -0.992657 -1.310147\n",
      "26  -0.623342 -1.310147\n",
      "27  -0.087835 -1.310147\n",
      "28  -0.992657  0.047235\n",
      "29  -0.300191  0.047235\n",
      "..        ...       ...\n",
      "516  0.392275  0.047235\n",
      "517  0.392275  1.404617\n",
      "518  0.503069  1.404617\n",
      "519  0.392275  0.047235\n",
      "520  1.167836  1.404617\n",
      "521  0.392275  1.404617\n",
      "522  0.392275  1.404617\n",
      "523  0.392275  1.404617\n",
      "524  0.392275  1.404617\n",
      "525  1.073199  0.047235\n",
      "526 -0.000123  0.047235\n",
      "527  0.542309  0.047235\n",
      "528  0.392275  0.047235\n",
      "529  0.392275  0.047235\n",
      "530  0.392275  0.047235\n",
      "531  0.853918  0.047235\n",
      "532  3.577617  0.047235\n",
      "533  1.170606  0.047235\n",
      "534 -0.069369  0.047235\n",
      "535  0.299946  0.047235\n",
      "536  0.392275  0.047235\n",
      "537 -0.161698  0.047235\n",
      "538  0.623097  0.047235\n",
      "539  1.001644  1.404617\n",
      "540  0.634638  0.047235\n",
      "541 -0.161698  0.047235\n",
      "542  0.392275  0.047235\n",
      "543  0.392275  0.047235\n",
      "544  0.392275  0.047235\n",
      "545  0.392275  0.047235\n",
      "\n",
      "[546 rows x 2 columns]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>          <td>price</td>      <th>  R-squared:         </th> <td>   0.370</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.368</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   159.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Fri, 08 Nov 2019</td> <th>  Prob (F-statistic):</th> <td>2.34e-55</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>14:29:19</td>     <th>  Log-Likelihood:    </th> <td> -647.99</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>   546</td>      <th>  AIC:               </th> <td>   1300.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>   544</td>      <th>  BIC:               </th> <td>   1309.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "      <td></td>        <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>lotsize</th>  <td>    0.4910</td> <td>    0.034</td> <td>   14.278</td> <td> 0.000</td> <td>    0.423</td> <td>    0.559</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>bedrooms</th> <td>    0.2915</td> <td>    0.034</td> <td>    8.477</td> <td> 0.000</td> <td>    0.224</td> <td>    0.359</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>77.789</td> <th>  Durbin-Watson:     </th> <td>   1.193</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td> 146.854</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.833</td> <th>  Prob(JB):          </th> <td>1.29e-32</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 4.919</td> <th>  Cond. No.          </th> <td>    1.17</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                  price   R-squared:                       0.370\n",
       "Model:                            OLS   Adj. R-squared:                  0.368\n",
       "Method:                 Least Squares   F-statistic:                     159.9\n",
       "Date:                Fri, 08 Nov 2019   Prob (F-statistic):           2.34e-55\n",
       "Time:                        14:29:19   Log-Likelihood:                -647.99\n",
       "No. Observations:                 546   AIC:                             1300.\n",
       "Df Residuals:                     544   BIC:                             1309.\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "lotsize        0.4910      0.034     14.278      0.000       0.423       0.559\n",
       "bedrooms       0.2915      0.034      8.477      0.000       0.224       0.359\n",
       "==============================================================================\n",
       "Omnibus:                       77.789   Durbin-Watson:                   1.193\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              146.854\n",
       "Skew:                           0.833   Prob(JB):                     1.29e-32\n",
       "Kurtosis:                       4.919   Cond. No.                         1.17\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#def multireg_verify():\n",
    "import statsmodels.api as sm\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "scale = StandardScaler()\n",
    "my_data = pd.read_csv('Housing.csv') #read the data\n",
    "print (my_data.shape)\n",
    "x=my_data\n",
    "newwdf=x[x.columns[1:4]]\n",
    "newwdf = (newwdf - newwdf.mean())/newwdf.std()\n",
    "print(newwdf)\n",
    "X = newwdf[['lotsize', 'bedrooms']]\n",
    "y = newwdf['price']\n",
    "\n",
    "X[['lotsize', 'bedrooms']] = scale.fit_transform(X[['lotsize', 'bedrooms']].as_matrix())\n",
    "print (X)\n",
    "\n",
    "est = sm.OLS(y, X).fit()\n",
    "\n",
    "    \n",
    "#multireg_verify()\n",
    "est.summary()\n",
    "#scaled = scale.transform ([[0.7,.945,0.05]])\n",
    "#print (scaled)\n",
    "#predicted = est.predict(scaled[0])\n",
    "#print (predicted)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "multivariate linear regression for auto mpg data\n",
      "(392, 9)\n",
      "(392, 4)\n",
      "[[ 0.  0.  0.]]\n",
      "(392, 1)\n",
      "(392, 3)\n",
      "[[ -3.80647894e-18   3.80584501e-01   3.20693300e-01]]\n",
      "0.0652843020402\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Pawan\\AppData\\Local\\Enthought\\Canopy\\edm\\envs\\User\\lib\\site-packages\\matplotlib\\figure.py:402: UserWarning: matplotlib is currently using a non-GUI backend, so cannot show the figure\n",
      "  \"matplotlib is currently using a non-GUI backend, \"\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEVCAYAAADuAi4fAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2d5685ac2e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2d569149be0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#############################################################################################\n",
    "#######################################################################################\n",
    "def mult_reg2():\n",
    "    print(\"multivariate linear regression for auto mpg data\")\n",
    "\n",
    "    ###########################################################\n",
    "    #######adjusting the data##################################\n",
    "    #multivariate linear regression\n",
    "    my_datta = pd.read_csv('auto-mpg(multivariate).csv') #read the data\n",
    "    print (my_datta.shape)\n",
    "    my_datta.head()\n",
    "    x=my_datta\n",
    "    #print(x)\n",
    "    newwdf = x[x.columns[:4]]\n",
    "    #print(newdf)\n",
    "    newwdf = (newwdf - newwdf.mean())/newwdf.std()\n",
    "    print (newwdf.shape)\n",
    "    newwdf.head()\n",
    "\n",
    "    #setting the matrixes\n",
    "\n",
    "    y = newwdf.iloc[:,1:3]\n",
    "    theta = np.zeros([1,3])\n",
    "\n",
    "    print (theta)\n",
    "    X = my_datta.iloc[:,1:2].values #.values converts it from pandas.core.frame.DataFrame to numpy.ndarray\n",
    "\n",
    "    ones = np.ones([y.shape[0],1])\n",
    "    y = np.concatenate((ones,y),axis=1)\n",
    "    \n",
    "\n",
    "    X = (X - X.mean())/X.std()\n",
    "\n",
    "    #print(X)\n",
    "    print (X.shape)\n",
    "\n",
    "    #print (y)\n",
    "    print (y.shape)\n",
    "\n",
    "    def computeCost(X,y,theta):\n",
    "\n",
    "        tobesummed = np.power(((y @ theta.T)-X),2)\n",
    "\n",
    "        return np.sum(tobesummed)/(2 * len(y))\n",
    "    def gradientDescent(X,y,theta,iters,alpha):\n",
    "\n",
    "        cost = np.zeros(iters)\n",
    "\n",
    "        for i in range(iters):\n",
    "\n",
    "            theta = theta - (alpha/len(y)) * np.sum(y * (y @ theta.T - X), axis=0)\n",
    "\n",
    "            cost[i] = computeCost(X,y,theta)\n",
    "\n",
    "\n",
    "\n",
    "        return theta,cost\n",
    "\n",
    "\n",
    "\n",
    "    #set hyper parameters\n",
    "\n",
    "    alpha = 0.06\n",
    "\n",
    "    iters = 10\n",
    "\n",
    "\n",
    "\n",
    "    g,cost = gradientDescent(X,y,theta,iters,alpha)\n",
    "\n",
    "    print(g)\n",
    "\n",
    "\n",
    "\n",
    "    finalCost = computeCost(X,y,g)\n",
    "\n",
    "    print(finalCost)\n",
    "\n",
    "\n",
    "\n",
    "    %matplotlib inline\n",
    "    my_data = pd.read_csv('auto-mpg(multivariate).csv') #read the data\n",
    "    #print (my_data.shape)\n",
    "    #my_data.head()\n",
    "    x=my_data\n",
    "    #print(x)\n",
    "\n",
    "\n",
    "    newdf = x[x.columns[:5]]\n",
    "    #print(newdf)\n",
    "    displacement=np.array(newdf.iloc[:,1])\n",
    "    #print (displacement)\n",
    "    mpg = np.array(newdf.iloc[:,0])\n",
    "    #print (mpg)\n",
    "    horsepower = np.array(newdf.iloc[:,2])\n",
    "    #print (horsepower)\n",
    "    weight = np.array(newdf.iloc[:,3])\n",
    "    acceleration = np.array(newdf.iloc[:,4]) \n",
    "    \n",
    "    \n",
    "    #Y=data['price'].values\n",
    "    #X=data['lotsize'].values\n",
    "    \n",
    "    fig,(ax1,ax2)=plt.subplots(1,2)\n",
    "    fig.suptitle('multivariate regression')\n",
    "    ax1.scatter(mpg,displacement,color='m', label='regression line')\n",
    "    ax2.scatter(mpg,horsepower,color ='b', label ='regresssion line' )\n",
    "    #plt.plot(x,y,color='m', label='regression line')\n",
    "    fig.show()\n",
    "\n",
    "    x=range(60)\n",
    "    y=range(0,5000)\n",
    "    fig=plt.figure()\n",
    "    ax1=fig.add_subplot(111)\n",
    "    ax1.scatter(mpg,displacement,s=15,color='m', label='mpg vs displacement')\n",
    "    ax1.scatter(mpg,horsepower,s=15, color ='b', alpha =1,label ='mpg vs horsepower' )\n",
    "    #ax1.scatter(mpg,weight,color ='y', label ='mpg vs weight')\n",
    "    ax1.scatter(mpg,acceleration,s=15,color ='g', label ='mpg vs acceleration' )\n",
    "    plt.legend(loc='upper right');\n",
    "    plt.show()\n",
    "\n",
    "            \n",
    "mult_reg2()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(392, 9)\n",
      "          mpg  displacement  horsepower\n",
      "0   -0.697747      1.075915    0.663285\n",
      "1   -1.082115      1.486832    1.572585\n",
      "2   -0.697747      1.181033    1.182885\n",
      "3   -0.953992      1.047246    1.182885\n",
      "4   -0.825870      1.028134    0.923085\n",
      "5   -1.082115      2.241772    2.429924\n",
      "6   -1.210238      2.480677    3.001484\n",
      "7   -1.210238      2.346890    2.871584\n",
      "8   -1.210238      2.490234    3.131384\n",
      "9   -1.082115      1.869080    2.222085\n",
      "10  -1.082115      1.802186    1.702485\n",
      "11  -1.210238      1.391269    1.442685\n",
      "12  -1.082115      1.964642    1.182885\n",
      "13  -1.210238      2.490234    3.131384\n",
      "14   0.070991     -0.777990   -0.246015\n",
      "15  -0.185255      0.034288   -0.246015\n",
      "16  -0.697747      0.043844   -0.194055\n",
      "17  -0.313378      0.053400   -0.505815\n",
      "18   0.455359     -0.930889   -0.427875\n",
      "19   0.327236     -0.930889   -1.519034\n",
      "20   0.199113     -0.806659   -0.453855\n",
      "21   0.070991     -0.835327   -0.375915\n",
      "22   0.199113     -0.863996   -0.246015\n",
      "23   0.327236     -0.701540    0.221625\n",
      "24  -0.313378      0.043844   -0.375915\n",
      "25  -1.722730      1.582394    2.871584\n",
      "26  -1.722730      1.075915    2.481884\n",
      "27  -1.594607      1.181033    2.741684\n",
      "28  -1.850853      1.047246    2.300025\n",
      "29   0.455359     -0.930889   -0.427875\n",
      "..        ...           ...         ...\n",
      "362  0.583482     -0.787546   -0.427875\n",
      "363  0.455359     -0.787546   -0.427875\n",
      "364  1.352219     -0.787546   -0.427875\n",
      "365  0.967851     -0.787546   -0.505815\n",
      "366  0.711605     -0.567753   -0.531795\n",
      "367  0.455359     -0.414854   -0.375915\n",
      "368  0.070991     -0.519972   -0.323955\n",
      "369  1.608465     -0.854440   -0.791594\n",
      "370  1.736588     -0.988227   -0.947474\n",
      "371  0.967851     -0.988227   -0.947474\n",
      "372  1.864711     -0.854440   -1.077374\n",
      "373  1.608465     -0.921333   -0.895514\n",
      "374  1.608465     -0.711097   -0.427875\n",
      "375  1.608465     -0.835327   -0.765614\n",
      "376  1.352219     -0.825771   -0.895514\n",
      "377  1.864711     -0.988227   -0.973454\n",
      "378  1.095974     -0.988227   -0.973454\n",
      "379  1.864711     -0.988227   -0.973454\n",
      "380  0.199113     -0.128168    0.143685\n",
      "381  1.864711      0.645885   -0.505815\n",
      "382  0.327236     -0.367073   -0.323955\n",
      "383 -0.185255      0.359199    0.195645\n",
      "384  1.095974     -0.481748   -0.220035\n",
      "385  1.608465     -0.567753   -0.531795\n",
      "386  0.455359     -0.414854   -0.375915\n",
      "387  0.455359     -0.519972   -0.479835\n",
      "388  2.633448     -0.930889   -1.363154\n",
      "389  1.095974     -0.567753   -0.531795\n",
      "390  0.583482     -0.711097   -0.661694\n",
      "391  0.967851     -0.720653   -0.583754\n",
      "\n",
      "[392 rows x 3 columns]\n",
      "     displacement  horsepower\n",
      "0        1.077290    0.664133\n",
      "1        1.488732    1.574594\n",
      "2        1.182542    1.184397\n",
      "3        1.048584    1.184397\n",
      "4        1.029447    0.924265\n",
      "5        2.244637    2.433030\n",
      "6        2.483848    3.005320\n",
      "7        2.349890    2.875254\n",
      "8        2.493416    3.135386\n",
      "9        1.871469    2.224924\n",
      "10       1.804490    1.704660\n",
      "11       1.393047    1.444529\n",
      "12       1.967153    1.184397\n",
      "13       2.493416    3.135386\n",
      "14      -0.778984   -0.246329\n",
      "15       0.034332   -0.246329\n",
      "16       0.043900   -0.194303\n",
      "17       0.053468   -0.506461\n",
      "18      -0.932079   -0.428421\n",
      "19      -0.932079   -1.520975\n",
      "20      -0.807690   -0.454435\n",
      "21      -0.836395   -0.376395\n",
      "22      -0.865100   -0.246329\n",
      "23      -0.702437    0.221908\n",
      "24       0.043900   -0.376395\n",
      "25       1.584416    2.875254\n",
      "26       1.077290    2.485056\n",
      "27       1.182542    2.745188\n",
      "28       1.048584    2.302964\n",
      "29      -0.932079   -0.428421\n",
      "..            ...         ...\n",
      "362     -0.788553   -0.428421\n",
      "363     -0.788553   -0.428421\n",
      "364     -0.788553   -0.428421\n",
      "365     -0.788553   -0.506461\n",
      "366     -0.568479   -0.532474\n",
      "367     -0.415384   -0.376395\n",
      "368     -0.520637   -0.324369\n",
      "369     -0.855532   -0.792606\n",
      "370     -0.989490   -0.948685\n",
      "371     -0.989490   -0.948685\n",
      "372     -0.855532   -1.078751\n",
      "373     -0.922511   -0.896659\n",
      "374     -0.712005   -0.428421\n",
      "375     -0.836395   -0.766593\n",
      "376     -0.826826   -0.896659\n",
      "377     -0.989490   -0.974698\n",
      "378     -0.989490   -0.974698\n",
      "379     -0.989490   -0.974698\n",
      "380     -0.128332    0.143869\n",
      "381      0.646711   -0.506461\n",
      "382     -0.367542   -0.324369\n",
      "383      0.359658    0.195895\n",
      "384     -0.482363   -0.220316\n",
      "385     -0.568479   -0.532474\n",
      "386     -0.415384   -0.376395\n",
      "387     -0.520637   -0.480448\n",
      "388     -0.932079   -1.364896\n",
      "389     -0.568479   -0.532474\n",
      "390     -0.712005   -0.662540\n",
      "391     -0.721574   -0.584501\n",
      "\n",
      "[392 rows x 2 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Pawan\\AppData\\Local\\Enthought\\Canopy\\edm\\envs\\User\\lib\\site-packages\\ipykernel\\__main__.py:13: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "C:\\Users\\Pawan\\AppData\\Local\\Enthought\\Canopy\\edm\\envs\\User\\lib\\site-packages\\pandas\\core\\frame.py:2356: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self.loc._setitem_with_indexer((slice(None), indexer), value)\n",
      "C:\\Users\\Pawan\\AppData\\Local\\Enthought\\Canopy\\edm\\envs\\User\\lib\\site-packages\\pandas\\core\\frame.py:2326: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self._setitem_array(key, value)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>           <td>mpg</td>       <th>  R-squared:         </th> <td>   0.664</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.663</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   385.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Fri, 08 Nov 2019</td> <th>  Prob (F-statistic):</th> <td>3.57e-93</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>14:29:21</td>     <th>  Log-Likelihood:    </th> <td> -341.76</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>   392</td>      <th>  AIC:               </th> <td>   687.5</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>   390</td>      <th>  BIC:               </th> <td>   695.5</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td></td>          <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>displacement</th> <td>   -0.5466</td> <td>    0.066</td> <td>   -8.236</td> <td> 0.000</td> <td>   -0.677</td> <td>   -0.416</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>horsepower</th>   <td>   -0.2870</td> <td>    0.066</td> <td>   -4.325</td> <td> 0.000</td> <td>   -0.417</td> <td>   -0.157</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>36.284</td> <th>  Durbin-Watson:     </th> <td>   0.931</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  44.828</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.736</td> <th>  Prob(JB):          </th> <td>1.84e-10</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.760</td> <th>  Cond. No.          </th> <td>    4.30</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                    mpg   R-squared:                       0.664\n",
       "Model:                            OLS   Adj. R-squared:                  0.663\n",
       "Method:                 Least Squares   F-statistic:                     385.9\n",
       "Date:                Fri, 08 Nov 2019   Prob (F-statistic):           3.57e-93\n",
       "Time:                        14:29:21   Log-Likelihood:                -341.76\n",
       "No. Observations:                 392   AIC:                             687.5\n",
       "Df Residuals:                     390   BIC:                             695.5\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "================================================================================\n",
       "                   coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------\n",
       "displacement    -0.5466      0.066     -8.236      0.000      -0.677      -0.416\n",
       "horsepower      -0.2870      0.066     -4.325      0.000      -0.417      -0.157\n",
       "==============================================================================\n",
       "Omnibus:                       36.284   Durbin-Watson:                   0.931\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               44.828\n",
       "Skew:                           0.736   Prob(JB):                     1.84e-10\n",
       "Kurtosis:                       3.760   Cond. No.                         4.30\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import statsmodels.api as sm\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "scale = StandardScaler()\n",
    "my_data = pd.read_csv('auto-mpg(multivariate).csv') #read the data\n",
    "print (my_data.shape)\n",
    "x=my_data\n",
    "nexdf=x[x.columns[:3]]\n",
    "nexdf = (nexdf - nexdf.mean())/nexdf.std()\n",
    "print(nexdf)\n",
    "X = nexdf[['displacement', 'horsepower']]\n",
    "y = nexdf['mpg']\n",
    "\n",
    "X[['displacement', 'horsepower']] = scale.fit_transform(X[['displacement', 'horsepower']].as_matrix())\n",
    "\n",
    "print (X)\n",
    "\n",
    "est = sm.OLS(y, X).fit()\n",
    "\n",
    "est.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 0.00149031  1.        ]\n",
      " [-0.2071535  -1.        ]\n",
      " [-0.30551416  1.        ]\n",
      " ..., \n",
      " [-0.21907601 -1.        ]\n",
      " [-0.10283159  1.        ]\n",
      " [-0.09388972 -1.        ]]\n",
      "(768, 2)\n",
      "768\n",
      "(567, 2)\n",
      "(567, 2)\n",
      "(768,)\n",
      "(567,)\n",
      "(567, 2)\n",
      "567\n",
      "Accuracy: 100.0%\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Pawan\\AppData\\Local\\Enthought\\Canopy\\edm\\envs\\User\\lib\\site-packages\\ipykernel\\__main__.py:66: RuntimeWarning: overflow encountered in exp\n"
     ]
    }
   ],
   "source": [
    "################################################\n",
    "#####################logistic regression################\n",
    "def log_reg():\n",
    "    import math\n",
    "\n",
    "    import numpy as np\n",
    "\n",
    "    import pandas as pd\n",
    "\n",
    "    from pandas import DataFrame\n",
    "\n",
    "    from sklearn import preprocessing\n",
    "\n",
    "    df = pd.read_csv(\"diabetes2(logistic).csv\")\n",
    "\n",
    "    # clean up data\n",
    "\n",
    "    # df.columns = [\"grade1\",\"grade2\",\"label\"]\n",
    "\n",
    "    x = df[\"Outcome\"].map(lambda x: float(str(x).rstrip(';')))\n",
    "\n",
    "    # formats the input data into two arrays, one of independant variables\n",
    "\n",
    "    # and one of the dependant variable\n",
    "\n",
    "    X = df[[\"BMI\",\"Outcome\"]]\n",
    "\n",
    "    X = np.array(X)\n",
    "\n",
    "    min_max_scaler = preprocessing.MinMaxScaler(feature_range=(-1,1))\n",
    "\n",
    "    X = min_max_scaler.fit_transform(X)\n",
    "\n",
    "    Y = df[\"BMI\"].map(lambda x: float(str(x).rstrip(';')))\n",
    "\n",
    "    Y = np.array(Y)\n",
    "\n",
    "    df.head()\n",
    "\n",
    "    print (X)\n",
    "\n",
    "    print (X.shape)\n",
    "    #X_train= np.splice(X)\n",
    "    type(X)\n",
    "    print(len(X))\n",
    "    #X_train=np.split(X,2)\n",
    "    X_train=X[:567]\n",
    "    #X_train\n",
    "    print(X_train.shape)\n",
    "    X_test =X[201:]\n",
    "    #X_test\n",
    "    print(X_test.shape)\n",
    "\n",
    "    #print (Y)\n",
    "    print (Y.shape)\n",
    "    type(Y)\n",
    "    Y_train =Y[:567]\n",
    "    #print (Y_train)\n",
    "    print(Y_train.shape)\n",
    "    Y_test =X[201:]\n",
    "    #Y_test\n",
    "    print(Y_test.shape)\n",
    "\n",
    "    def logistic_func(z):\n",
    "\n",
    "       pr_y = float(1.0 / float((1.0 + np.exp(-1.0*z))))\n",
    "\n",
    "       return pr_y\n",
    "\n",
    "\n",
    "\n",
    "    def calc_hypothesis(theta, x):\n",
    "\n",
    "       z = 0\n",
    "\n",
    "       for i in range(len(theta)):\n",
    "\n",
    "          z += x[i]*theta[i]\n",
    "\n",
    "       return logistic_func(z)\n",
    "\n",
    "    def Cost_Function(X,Y,theta,m):\n",
    "\n",
    "       sumOfErrors = 0\n",
    "\n",
    "       for i in range(m):\n",
    "\n",
    "          xi = X[i]\n",
    "\n",
    "          hi = calc_hypothesis(theta,xi)\n",
    "\n",
    "          if Y[i] == 1:\n",
    "\n",
    "             error = Y[i] * math.log(hi)\n",
    "\n",
    "          elif Y[i] == 0:\n",
    "\n",
    "             error = (1-Y[i]) * math.log(1-hi)\n",
    "\n",
    "          sumOfErrors += error\n",
    "\n",
    "       const = -1/m\n",
    "\n",
    "       J = const * sumOfErrors\n",
    "\n",
    "       print ('cost is ', J)\n",
    "\n",
    "       return J\n",
    "    def partial_derivate_cost(X,Y,theta,j,m,alpha):\n",
    "\n",
    "       errors_sum = 0\n",
    "\n",
    "       for i in range(m):\n",
    "\n",
    "          xi = X[i]\n",
    "\n",
    "          xij = xi[j]\n",
    "\n",
    "          hi = calc_hypothesis(theta,X[i])\n",
    "\n",
    "          error = (hi - Y[i])*xij # partial derivative w.r.t xij\n",
    "\n",
    "          errors_sum += error\n",
    "\n",
    "       m = len(Y)\n",
    "\n",
    "       constant = float(alpha)/float(m)\n",
    "\n",
    "       J = constant * errors_sum\n",
    "\n",
    "       return J\n",
    "    def gradient_descent(X,Y,theta,m,alpha):\n",
    "\n",
    "       theta_new = []\n",
    "\n",
    "       for pos_i in range(len(theta)):\n",
    "\n",
    "          CFDerivative = partial_derivate_cost(X,Y,theta,pos_i,m,alpha)\n",
    "\n",
    "          updated_theta = theta[pos_i] - CFDerivative\n",
    "\n",
    "          theta_new.append(updated_theta)\n",
    "\n",
    "       return theta_new\n",
    "    def logistic_regression(X,Y,alpha,theta,num_iters):\n",
    "\n",
    "       m = len(Y)\n",
    "\n",
    "       for x in range(num_iters):\n",
    "\n",
    "          new_theta = gradient_descent(X,Y,theta,m,alpha)\n",
    "\n",
    "          theta = new_theta\n",
    "\n",
    "       return theta \n",
    "    initial_theta = [0,0] # Initial guess\n",
    "\n",
    "    alpha = 0.1 # learning rate\n",
    "\n",
    "    iterations = 1000 # Number of iterations\n",
    "\n",
    "    optimal_theta = logistic_regression(X_train,Y_train,alpha,initial_theta,iterations)\n",
    "\n",
    "    # Testing the model for accuracy\n",
    "\n",
    "    score = 0\n",
    "\n",
    "    length_test = len(X_test)\n",
    "\n",
    "    for i in range(length_test):\n",
    "\n",
    "       prediction = round(calc_hypothesis(X_test[i],optimal_theta))\n",
    "\n",
    "       answer = Y_test[i]\n",
    "\n",
    "       if (prediction==answer).all:\n",
    "\n",
    "          score += 1\n",
    "\n",
    "    #the same process is repeated for the implementation from this module and the scores compared to find the higher match-rate\n",
    "    print (score)\n",
    "    my_score = float(score) / float(length_test)\n",
    "\n",
    "    print('Accuracy: {0}%'.format(round(my_score*100, 2)))\n",
    "log_reg()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pregnancies</th>\n",
       "      <th>Glucose</th>\n",
       "      <th>BloodPressure</th>\n",
       "      <th>SkinThickness</th>\n",
       "      <th>Insulin</th>\n",
       "      <th>BMI</th>\n",
       "      <th>DiabetesPedigreeFunction</th>\n",
       "      <th>Age</th>\n",
       "      <th>Outcome</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6</td>\n",
       "      <td>148</td>\n",
       "      <td>72</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>33.6</td>\n",
       "      <td>0.627</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>85</td>\n",
       "      <td>66</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>26.6</td>\n",
       "      <td>0.351</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8</td>\n",
       "      <td>183</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>23.3</td>\n",
       "      <td>0.672</td>\n",
       "      <td>32</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>89</td>\n",
       "      <td>66</td>\n",
       "      <td>23</td>\n",
       "      <td>94</td>\n",
       "      <td>28.1</td>\n",
       "      <td>0.167</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>137</td>\n",
       "      <td>40</td>\n",
       "      <td>35</td>\n",
       "      <td>168</td>\n",
       "      <td>43.1</td>\n",
       "      <td>2.288</td>\n",
       "      <td>33</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pregnancies  Glucose  BloodPressure  SkinThickness  Insulin   BMI  \\\n",
       "0            6      148             72             35        0  33.6   \n",
       "1            1       85             66             29        0  26.6   \n",
       "2            8      183             64              0        0  23.3   \n",
       "3            1       89             66             23       94  28.1   \n",
       "4            0      137             40             35      168  43.1   \n",
       "\n",
       "   DiabetesPedigreeFunction  Age  Outcome  \n",
       "0                     0.627   50        1  \n",
       "1                     0.351   31        0  \n",
       "2                     0.672   32        1  \n",
       "3                     0.167   21        0  \n",
       "4                     2.288   33        1  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"diabetes2(logistic).csv\")\n",
    "x = df[\"Outcome\"].map(lambda x: float(str(x).rstrip(';')))\n",
    "X = df[[\"BMI\",\"Outcome\"]]\n",
    "\n",
    "X = np.array(X)\n",
    "min_max_scaler = preprocessing.MinMaxScaler(feature_range=(-1,1))\n",
    "\n",
    "X = min_max_scaler.fit_transform(X)\n",
    "\n",
    "Y = df[\"BMI\"].map(lambda x: float(str(x).rstrip(';')))\n",
    "\n",
    "Y = np.array(Y)\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(768, 9)\n",
      "(768, 9)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pregnancies</th>\n",
       "      <th>Glucose</th>\n",
       "      <th>BloodPressure</th>\n",
       "      <th>SkinThickness</th>\n",
       "      <th>Insulin</th>\n",
       "      <th>BMI</th>\n",
       "      <th>DiabetesPedigreeFunction</th>\n",
       "      <th>Age</th>\n",
       "      <th>Outcome</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6</td>\n",
       "      <td>148</td>\n",
       "      <td>72</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>33.6</td>\n",
       "      <td>0.627</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>85</td>\n",
       "      <td>66</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>26.6</td>\n",
       "      <td>0.351</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8</td>\n",
       "      <td>183</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>23.3</td>\n",
       "      <td>0.672</td>\n",
       "      <td>32</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>89</td>\n",
       "      <td>66</td>\n",
       "      <td>23</td>\n",
       "      <td>94</td>\n",
       "      <td>28.1</td>\n",
       "      <td>0.167</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>137</td>\n",
       "      <td>40</td>\n",
       "      <td>35</td>\n",
       "      <td>168</td>\n",
       "      <td>43.1</td>\n",
       "      <td>2.288</td>\n",
       "      <td>33</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pregnancies  Glucose  BloodPressure  SkinThickness  Insulin   BMI  \\\n",
       "0            6      148             72             35        0  33.6   \n",
       "1            1       85             66             29        0  26.6   \n",
       "2            8      183             64              0        0  23.3   \n",
       "3            1       89             66             23       94  28.1   \n",
       "4            0      137             40             35      168  43.1   \n",
       "\n",
       "   DiabetesPedigreeFunction  Age  Outcome  \n",
       "0                     0.627   50        1  \n",
       "1                     0.351   31        0  \n",
       "2                     0.672   32        1  \n",
       "3                     0.167   21        0  \n",
       "4                     2.288   33        1  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAD8CAYAAAB9y7/cAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2d568a5f860>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mydata = pd.read_csv('diabetes2(logistic).csv') #read the data\n",
    "#print (mydata.shape)\n",
    "mydata.head()\n",
    "x=mydata\n",
    "    #print(x)\n",
    "ndf = x[x.columns[:9]]\n",
    "bmi=np.array(ndf.iloc[:,5])\n",
    "outcome =np.array(ndf.iloc[:,8])\n",
    "#print (outcome)\n",
    "plt.scatter(bmi,outcome)\n",
    "#print(theta_new)\n",
    "#plt.plot(bmi,qqqqq )\n",
    "print (ndf.shape)\n",
    "ndf.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.77489177489177485"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(\"diabetes2(logistic).csv\") #importing the dataset\n",
    "data.sample(5)\n",
    "X = data.iloc[:,0:-1] # All columns except last one as X\n",
    "y = data.iloc[:,-1] # Only last column as y\n",
    "\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=.30,random_state=1) #splitting the data as train and test\n",
    "X_train.shape\n",
    "X_test.shape\n",
    "\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "l_reg = LogisticRegression() # Making a logistic regression model\n",
    "l_reg.fit(X_train,y_train) # Fitting the data\n",
    "y_pred = l_reg.predict(X_test) # Predict the X_test data\n",
    "from sklearn import metrics\n",
    "metrics.accuracy_score(y_test,y_pred) # calculate the accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(25, 26)\n",
      "[-0.54298643 -0.63348416 -0.723981   -0.81447964 -0.90314    -1.0859728\n",
      " -1.2669683  -1.4479638  -1.62895928 -1.665158   -1.701357   -1.73755656\n",
      " -1.809954   -1.99095    -2.22       -2.208      -2.2443439  -2.28054299\n",
      " -2.31674208 -2.35294118 -2.38914027 -2.42533937 -2.46153846 -2.49773756\n",
      " -2.53393665]\n",
      "[ -3.   -3.5  -4.    4.5  -5.   -6.   -7.   -8.   -9.   -9.2  -9.4  -9.6\n",
      " -10.  -11.  -12.  -12.2 -12.4 -12.6 -12.8 -13.  -13.2 -13.4 -13.6 -13.8\n",
      " -14. ]\n",
      "[  -8168.4849  -11246.7262  -14852.882   -18957.0485  -23755.818\n",
      "  -34379.928   -47262.674   -62246.219   -79525.027   -83199.644\n",
      "  -86859.126   -90709.414   -98538.738  -120017.5    -132259.48   -148368.07\n",
      " -153418.31   -158598.22   -163710.27   -169063.32   -174443.88   -179922.75\n",
      " -185392.35   -191027.4    -196746.12  ]\n",
      "[ -24505.4547  -33740.1786  -44558.646   -56871.1455  -71267.454\n",
      " -103139.784  -141788.022  -186738.657  -238575.081  -249598.932\n",
      " -260577.378  -272128.242  -295616.214  -360052.5    -396778.44   -445104.21\n",
      " -460254.93   -475794.66   -491130.81   -507189.96   -523331.64   -539768.25\n",
      " -556177.05   -573082.2    -590238.36  ]\n",
      "[   13306.12925    21373.8688     32259.65319    46320.39       64364.48841\n",
      "   112007.        179640.9292    270390.8156    388629.0912    415621.6584\n",
      "   443335.1461    472838.2123    535051.749     716846.5249    880848.1368\n",
      "   982790.0957   1032970.345    1085070.175    1137823.415    1193388.141\n",
      "  1250312.696    1309121.186    1369051.2      1431408.934    1495626.61   ]\n",
      "[ 0.49825495  7.90687689  2.80896631]\n",
      "[ -1.75490618  -2.5155516   -3.27618944  -4.03684243  -4.78204446\n",
      "  -6.31877824  -7.84006928  -9.36136033 -10.88265116 -11.18690621\n",
      " -11.49116357 -11.79542566 -12.40393568 -13.92523092 -15.85042636\n",
      " -15.74956478 -16.05504005 -16.35929814 -16.66355631 -16.96781448\n",
      " -17.27207264 -17.57633082 -17.88058898 -18.18484715 -18.48910531]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJgAAAJQCAYAAADCP95TAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2d568be5d30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "'1e+01'"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "####################################################################################\n",
    "################################Polynomial Regression##############################\n",
    "plt.rcParams['figure.figsize']=(20.0,10.0)\n",
    "data = pd.read_csv('power gen eff data turbines (poly) reg.csv')\n",
    "print(data.shape)\n",
    "data.head()\n",
    "blah_x=data['Rotvel'].values\n",
    "blah_y=data['Vz'].values\n",
    "blah_z=data['ST'].values\n",
    "blah_zz=data['TT'].values\n",
    "blah_pp=data['Power'].values\n",
    "\n",
    "print(blah_x)\n",
    "print(blah_y)\n",
    "print(blah_z)\n",
    "print(blah_zz)\n",
    "\n",
    "print(blah_pp)\n",
    "coeffs = np.polyfit(blah_x,blah_y,deg=2)\n",
    "print (coeffs)\n",
    "\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "\n",
    "#x = np.arange(0, -15, 0.25)\n",
    "#y = np.arange(0, -15, 0.25)\n",
    "X, Y = np.meshgrid(blah_x,blah_y)\n",
    "F = (np.multiply(coeffs[0],blah_x))+(np.multiply(coeffs[1],blah_x))+coeffs[2]\n",
    "#F2= (np.multiply(coeffs[0],blah_y))+(np.multiply(coeffs[1],blah_y))+coeffs[2]\n",
    "print (F)\n",
    "blah_x.shape,blah_y.shape,blah_z.shape,blah_pp.shape,coeffs.shape, F.shape\n",
    "\n",
    "#3 + 2*X + 4*X*Y + 5*X*X)\n",
    "fig =plt.figure()\n",
    "ax1 = fig.add_subplot(111)\n",
    "plt.xlabel('Rot vel')\n",
    "plt.ylabel('Vz')\n",
    "ax1.scatter(blah_x,blah_y, label='scatter plot')\n",
    "ax1.plot(blah_x,F, c='r', label = 'Poly line' ) #the line is actually a part of polynomial of 2nd degree.\n",
    "#since the scatter plot demands that we plot a  miniscule area we end up getting a plot that looks like a line. \n",
    "\n",
    "plt.legend(loc='upper left');\n",
    "#ax1.plot(blah_y,F2, c='y')\n",
    "plt.show()\n",
    "yf = np.polyval(np.poly1d(coeffs), blah_x)\n",
    "yf.shape\n",
    "yf\n",
    "'%.1g' % max(blah_x-yf)                                   \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}