wine.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Relevant Information:\n",
    "\n",
    "       -- These data are the results of a chemical analysis of\n",
    "          wines grown in the same region in Italy but derived from three\n",
    "          different cultivars.\n",
    "          The analysis determined the quantities of 13 constituents\n",
    "          found in each of the three types of wines. \n",
    "\n",
    "\n",
    "       -- The attributes are:\n",
    "        1) Alcohol\n",
    "        2) Malic acid\n",
    "        3) Ash\n",
    "        4) Alcalinity of ash  \n",
    "        5) Magnesium\n",
    "        6) Total phenols\n",
    "        7) Flavanoids\n",
    "        8) Nonflavanoid phenols\n",
    "        9) Proanthocyanins\n",
    "        10)Color intensity\n",
    "        11)Hue\n",
    "        12)OD280/OD315 of diluted wines\n",
    "        13)Proline            \n",
    "\n",
    "* Number of Instances\n",
    "\n",
    "        class 1 59\n",
    "        class 2 71\n",
    "        class 3 48\n",
    "\n",
    "* Number of Attributes \n",
    "\t\n",
    "        13\n",
    "\n",
    "* For Each Attribute:\n",
    "\n",
    "        All attributes are continuous\n",
    "\t\n",
    "        No statistics available, but suggest to standardise\n",
    "        variables for certain uses (e.g. for us with classifiers\n",
    "        which are NOT scale invariant)\n",
    "\n",
    "        NOTE: 1st attribute is class identifier (1-3)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>class</th>\n",
       "      <th>Alcohol</th>\n",
       "      <th>Malic acid</th>\n",
       "      <th>Ash</th>\n",
       "      <th>Alcalinity of ash</th>\n",
       "      <th>Magnesium</th>\n",
       "      <th>Total phenols</th>\n",
       "      <th>Flavanoids</th>\n",
       "      <th>Nonflavanoid phenols</th>\n",
       "      <th>Proanthocyanins</th>\n",
       "      <th>Color intensity</th>\n",
       "      <th>Hue</th>\n",
       "      <th>OD280/OD315 of diluted wines</th>\n",
       "      <th>Proline</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>14.23</td>\n",
       "      <td>1.71</td>\n",
       "      <td>2.43</td>\n",
       "      <td>15.6</td>\n",
       "      <td>127</td>\n",
       "      <td>2.80</td>\n",
       "      <td>3.06</td>\n",
       "      <td>0.28</td>\n",
       "      <td>2.29</td>\n",
       "      <td>5.64</td>\n",
       "      <td>1.04</td>\n",
       "      <td>3.92</td>\n",
       "      <td>1065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>13.20</td>\n",
       "      <td>1.78</td>\n",
       "      <td>2.14</td>\n",
       "      <td>11.2</td>\n",
       "      <td>100</td>\n",
       "      <td>2.65</td>\n",
       "      <td>2.76</td>\n",
       "      <td>0.26</td>\n",
       "      <td>1.28</td>\n",
       "      <td>4.38</td>\n",
       "      <td>1.05</td>\n",
       "      <td>3.40</td>\n",
       "      <td>1050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>13.16</td>\n",
       "      <td>2.36</td>\n",
       "      <td>2.67</td>\n",
       "      <td>18.6</td>\n",
       "      <td>101</td>\n",
       "      <td>2.80</td>\n",
       "      <td>3.24</td>\n",
       "      <td>0.30</td>\n",
       "      <td>2.81</td>\n",
       "      <td>5.68</td>\n",
       "      <td>1.03</td>\n",
       "      <td>3.17</td>\n",
       "      <td>1185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>14.37</td>\n",
       "      <td>1.95</td>\n",
       "      <td>2.50</td>\n",
       "      <td>16.8</td>\n",
       "      <td>113</td>\n",
       "      <td>3.85</td>\n",
       "      <td>3.49</td>\n",
       "      <td>0.24</td>\n",
       "      <td>2.18</td>\n",
       "      <td>7.80</td>\n",
       "      <td>0.86</td>\n",
       "      <td>3.45</td>\n",
       "      <td>1480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>13.24</td>\n",
       "      <td>2.59</td>\n",
       "      <td>2.87</td>\n",
       "      <td>21.0</td>\n",
       "      <td>118</td>\n",
       "      <td>2.80</td>\n",
       "      <td>2.69</td>\n",
       "      <td>0.39</td>\n",
       "      <td>1.82</td>\n",
       "      <td>4.32</td>\n",
       "      <td>1.04</td>\n",
       "      <td>2.93</td>\n",
       "      <td>735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>173</th>\n",
       "      <td>3</td>\n",
       "      <td>13.71</td>\n",
       "      <td>5.65</td>\n",
       "      <td>2.45</td>\n",
       "      <td>20.5</td>\n",
       "      <td>95</td>\n",
       "      <td>1.68</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.52</td>\n",
       "      <td>1.06</td>\n",
       "      <td>7.70</td>\n",
       "      <td>0.64</td>\n",
       "      <td>1.74</td>\n",
       "      <td>740</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>174</th>\n",
       "      <td>3</td>\n",
       "      <td>13.40</td>\n",
       "      <td>3.91</td>\n",
       "      <td>2.48</td>\n",
       "      <td>23.0</td>\n",
       "      <td>102</td>\n",
       "      <td>1.80</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.43</td>\n",
       "      <td>1.41</td>\n",
       "      <td>7.30</td>\n",
       "      <td>0.70</td>\n",
       "      <td>1.56</td>\n",
       "      <td>750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>175</th>\n",
       "      <td>3</td>\n",
       "      <td>13.27</td>\n",
       "      <td>4.28</td>\n",
       "      <td>2.26</td>\n",
       "      <td>20.0</td>\n",
       "      <td>120</td>\n",
       "      <td>1.59</td>\n",
       "      <td>0.69</td>\n",
       "      <td>0.43</td>\n",
       "      <td>1.35</td>\n",
       "      <td>10.20</td>\n",
       "      <td>0.59</td>\n",
       "      <td>1.56</td>\n",
       "      <td>835</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>176</th>\n",
       "      <td>3</td>\n",
       "      <td>13.17</td>\n",
       "      <td>2.59</td>\n",
       "      <td>2.37</td>\n",
       "      <td>20.0</td>\n",
       "      <td>120</td>\n",
       "      <td>1.65</td>\n",
       "      <td>0.68</td>\n",
       "      <td>0.53</td>\n",
       "      <td>1.46</td>\n",
       "      <td>9.30</td>\n",
       "      <td>0.60</td>\n",
       "      <td>1.62</td>\n",
       "      <td>840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>177</th>\n",
       "      <td>3</td>\n",
       "      <td>14.13</td>\n",
       "      <td>4.10</td>\n",
       "      <td>2.74</td>\n",
       "      <td>24.5</td>\n",
       "      <td>96</td>\n",
       "      <td>2.05</td>\n",
       "      <td>0.76</td>\n",
       "      <td>0.56</td>\n",
       "      <td>1.35</td>\n",
       "      <td>9.20</td>\n",
       "      <td>0.61</td>\n",
       "      <td>1.60</td>\n",
       "      <td>560</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>178 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     class  Alcohol  Malic acid   Ash  Alcalinity of ash   Magnesium  \\\n",
       "0        1    14.23        1.71  2.43                15.6        127   \n",
       "1        1    13.20        1.78  2.14                11.2        100   \n",
       "2        1    13.16        2.36  2.67                18.6        101   \n",
       "3        1    14.37        1.95  2.50                16.8        113   \n",
       "4        1    13.24        2.59  2.87                21.0        118   \n",
       "..     ...      ...         ...   ...                 ...        ...   \n",
       "173      3    13.71        5.65  2.45                20.5         95   \n",
       "174      3    13.40        3.91  2.48                23.0        102   \n",
       "175      3    13.27        4.28  2.26                20.0        120   \n",
       "176      3    13.17        2.59  2.37                20.0        120   \n",
       "177      3    14.13        4.10  2.74                24.5         96   \n",
       "\n",
       "     Total phenols  Flavanoids  Nonflavanoid phenols  Proanthocyanins  \\\n",
       "0             2.80        3.06                  0.28             2.29   \n",
       "1             2.65        2.76                  0.26             1.28   \n",
       "2             2.80        3.24                  0.30             2.81   \n",
       "3             3.85        3.49                  0.24             2.18   \n",
       "4             2.80        2.69                  0.39             1.82   \n",
       "..             ...         ...                   ...              ...   \n",
       "173           1.68        0.61                  0.52             1.06   \n",
       "174           1.80        0.75                  0.43             1.41   \n",
       "175           1.59        0.69                  0.43             1.35   \n",
       "176           1.65        0.68                  0.53             1.46   \n",
       "177           2.05        0.76                  0.56             1.35   \n",
       "\n",
       "     Color intensity   Hue  OD280/OD315 of diluted wines  Proline  \n",
       "0               5.64  1.04                          3.92     1065  \n",
       "1               4.38  1.05                          3.40     1050  \n",
       "2               5.68  1.03                          3.17     1185  \n",
       "3               7.80  0.86                          3.45     1480  \n",
       "4               4.32  1.04                          2.93      735  \n",
       "..               ...   ...                           ...      ...  \n",
       "173             7.70  0.64                          1.74      740  \n",
       "174             7.30  0.70                          1.56      750  \n",
       "175            10.20  0.59                          1.56      835  \n",
       "176             9.30  0.60                          1.62      840  \n",
       "177             9.20  0.61                          1.60      560  \n",
       "\n",
       "[178 rows x 14 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#import Data\n",
    "\n",
    "names=\\\n",
    "[\"class\"\n",
    ",'Alcohol'\n",
    ",'Malic acid'\n",
    ",'Ash'\n",
    ",'Alcalinity of ash ' \n",
    ",'Magnesium'\n",
    ",'Total phenols'\n",
    ",'Flavanoids'\n",
    ",'Nonflavanoid phenols'\n",
    ",'Proanthocyanins'\n",
    ",'Color intensity'\n",
    ",'Hue'\n",
    ",'OD280/OD315 of diluted wines'\n",
    ",'Proline']\n",
    "\n",
    "data=pd.read_csv(\"wine.data\",names=names)\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>class</th>\n",
       "      <th>Alcohol</th>\n",
       "      <th>Malic acid</th>\n",
       "      <th>Ash</th>\n",
       "      <th>Alcalinity of ash</th>\n",
       "      <th>Magnesium</th>\n",
       "      <th>Total phenols</th>\n",
       "      <th>Flavanoids</th>\n",
       "      <th>Nonflavanoid phenols</th>\n",
       "      <th>Proanthocyanins</th>\n",
       "      <th>Color intensity</th>\n",
       "      <th>Hue</th>\n",
       "      <th>OD280/OD315 of diluted wines</th>\n",
       "      <th>Proline</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>125</th>\n",
       "      <td>2</td>\n",
       "      <td>12.07</td>\n",
       "      <td>2.16</td>\n",
       "      <td>2.17</td>\n",
       "      <td>21.0</td>\n",
       "      <td>85</td>\n",
       "      <td>2.60</td>\n",
       "      <td>2.65</td>\n",
       "      <td>0.37</td>\n",
       "      <td>1.35</td>\n",
       "      <td>2.76</td>\n",
       "      <td>0.86</td>\n",
       "      <td>3.28</td>\n",
       "      <td>378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90</th>\n",
       "      <td>2</td>\n",
       "      <td>12.08</td>\n",
       "      <td>1.83</td>\n",
       "      <td>2.32</td>\n",
       "      <td>18.5</td>\n",
       "      <td>81</td>\n",
       "      <td>1.60</td>\n",
       "      <td>1.50</td>\n",
       "      <td>0.52</td>\n",
       "      <td>1.64</td>\n",
       "      <td>2.40</td>\n",
       "      <td>1.08</td>\n",
       "      <td>2.27</td>\n",
       "      <td>480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>14.37</td>\n",
       "      <td>1.95</td>\n",
       "      <td>2.50</td>\n",
       "      <td>16.8</td>\n",
       "      <td>113</td>\n",
       "      <td>3.85</td>\n",
       "      <td>3.49</td>\n",
       "      <td>0.24</td>\n",
       "      <td>2.18</td>\n",
       "      <td>7.80</td>\n",
       "      <td>0.86</td>\n",
       "      <td>3.45</td>\n",
       "      <td>1480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102</th>\n",
       "      <td>2</td>\n",
       "      <td>12.34</td>\n",
       "      <td>2.45</td>\n",
       "      <td>2.46</td>\n",
       "      <td>21.0</td>\n",
       "      <td>98</td>\n",
       "      <td>2.56</td>\n",
       "      <td>2.11</td>\n",
       "      <td>0.34</td>\n",
       "      <td>1.31</td>\n",
       "      <td>2.80</td>\n",
       "      <td>0.80</td>\n",
       "      <td>3.38</td>\n",
       "      <td>438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>114</th>\n",
       "      <td>2</td>\n",
       "      <td>12.08</td>\n",
       "      <td>1.39</td>\n",
       "      <td>2.50</td>\n",
       "      <td>22.5</td>\n",
       "      <td>84</td>\n",
       "      <td>2.56</td>\n",
       "      <td>2.29</td>\n",
       "      <td>0.43</td>\n",
       "      <td>1.04</td>\n",
       "      <td>2.90</td>\n",
       "      <td>0.93</td>\n",
       "      <td>3.19</td>\n",
       "      <td>385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>3</td>\n",
       "      <td>12.70</td>\n",
       "      <td>3.55</td>\n",
       "      <td>2.36</td>\n",
       "      <td>21.5</td>\n",
       "      <td>106</td>\n",
       "      <td>1.70</td>\n",
       "      <td>1.20</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.84</td>\n",
       "      <td>5.00</td>\n",
       "      <td>0.78</td>\n",
       "      <td>1.29</td>\n",
       "      <td>600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>156</th>\n",
       "      <td>3</td>\n",
       "      <td>13.84</td>\n",
       "      <td>4.12</td>\n",
       "      <td>2.38</td>\n",
       "      <td>19.5</td>\n",
       "      <td>89</td>\n",
       "      <td>1.80</td>\n",
       "      <td>0.83</td>\n",
       "      <td>0.48</td>\n",
       "      <td>1.56</td>\n",
       "      <td>9.01</td>\n",
       "      <td>0.57</td>\n",
       "      <td>1.64</td>\n",
       "      <td>480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>1</td>\n",
       "      <td>13.71</td>\n",
       "      <td>1.86</td>\n",
       "      <td>2.36</td>\n",
       "      <td>16.6</td>\n",
       "      <td>101</td>\n",
       "      <td>2.61</td>\n",
       "      <td>2.88</td>\n",
       "      <td>0.27</td>\n",
       "      <td>1.69</td>\n",
       "      <td>3.80</td>\n",
       "      <td>1.11</td>\n",
       "      <td>4.00</td>\n",
       "      <td>1035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>113</th>\n",
       "      <td>2</td>\n",
       "      <td>11.41</td>\n",
       "      <td>0.74</td>\n",
       "      <td>2.50</td>\n",
       "      <td>21.0</td>\n",
       "      <td>88</td>\n",
       "      <td>2.48</td>\n",
       "      <td>2.01</td>\n",
       "      <td>0.42</td>\n",
       "      <td>1.44</td>\n",
       "      <td>3.08</td>\n",
       "      <td>1.10</td>\n",
       "      <td>2.31</td>\n",
       "      <td>434</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80</th>\n",
       "      <td>2</td>\n",
       "      <td>12.00</td>\n",
       "      <td>0.92</td>\n",
       "      <td>2.00</td>\n",
       "      <td>19.0</td>\n",
       "      <td>86</td>\n",
       "      <td>2.42</td>\n",
       "      <td>2.26</td>\n",
       "      <td>0.30</td>\n",
       "      <td>1.43</td>\n",
       "      <td>2.50</td>\n",
       "      <td>1.38</td>\n",
       "      <td>3.12</td>\n",
       "      <td>278</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>178 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     class  Alcohol  Malic acid   Ash  Alcalinity of ash   Magnesium  \\\n",
       "125      2    12.07        2.16  2.17                21.0         85   \n",
       "90       2    12.08        1.83  2.32                18.5         81   \n",
       "3        1    14.37        1.95  2.50                16.8        113   \n",
       "102      2    12.34        2.45  2.46                21.0         98   \n",
       "114      2    12.08        1.39  2.50                22.5         84   \n",
       "..     ...      ...         ...   ...                 ...        ...   \n",
       "133      3    12.70        3.55  2.36                21.5        106   \n",
       "156      3    13.84        4.12  2.38                19.5         89   \n",
       "22       1    13.71        1.86  2.36                16.6        101   \n",
       "113      2    11.41        0.74  2.50                21.0         88   \n",
       "80       2    12.00        0.92  2.00                19.0         86   \n",
       "\n",
       "     Total phenols  Flavanoids  Nonflavanoid phenols  Proanthocyanins  \\\n",
       "125           2.60        2.65                  0.37             1.35   \n",
       "90            1.60        1.50                  0.52             1.64   \n",
       "3             3.85        3.49                  0.24             2.18   \n",
       "102           2.56        2.11                  0.34             1.31   \n",
       "114           2.56        2.29                  0.43             1.04   \n",
       "..             ...         ...                   ...              ...   \n",
       "133           1.70        1.20                  0.17             0.84   \n",
       "156           1.80        0.83                  0.48             1.56   \n",
       "22            2.61        2.88                  0.27             1.69   \n",
       "113           2.48        2.01                  0.42             1.44   \n",
       "80            2.42        2.26                  0.30             1.43   \n",
       "\n",
       "     Color intensity   Hue  OD280/OD315 of diluted wines  Proline  \n",
       "125             2.76  0.86                          3.28      378  \n",
       "90              2.40  1.08                          2.27      480  \n",
       "3               7.80  0.86                          3.45     1480  \n",
       "102             2.80  0.80                          3.38      438  \n",
       "114             2.90  0.93                          3.19      385  \n",
       "..               ...   ...                           ...      ...  \n",
       "133             5.00  0.78                          1.29      600  \n",
       "156             9.01  0.57                          1.64      480  \n",
       "22              3.80  1.11                          4.00     1035  \n",
       "113             3.08  1.10                          2.31      434  \n",
       "80              2.50  1.38                          3.12      278  \n",
       "\n",
       "[178 rows x 14 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#shuffle data(Because the data is sorted by class)\n",
    "data=data.sample(axis=0,frac=1)\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(178, 13) (178,)\n"
     ]
    }
   ],
   "source": [
    "#convert to numpy array\n",
    "x=np.array(data[data.columns[1:14]])\n",
    "y=np.array(data[data.columns[0]])\n",
    "\n",
    "print(x.shape,y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of features = 13\n",
      "labels = [1 2 3]\n"
     ]
    }
   ],
   "source": [
    "print(f\"number of features = {x[0].shape[0]}\\nlabels = {np.unique(y)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "#data normalization\n",
    "mu=x.mean(axis=0)\n",
    "sigma=x.std(axis=0)\n",
    "\n",
    "x_norm=(x-mu)/sigma\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(124, 13)\n",
      "(124,)\n",
      "(54, 13)\n",
      "(54,)\n"
     ]
    }
   ],
   "source": [
    "x_train,x_test=x_norm[0:int(len(x)*.7)],x_norm[int(len(x)*.7):]\n",
    "y_train,y_test=y[0:int(len(x)*.7)],y[int(len(x)*.7):]\n",
    "\n",
    "print(x_train.shape)\n",
    "print(y_train.shape)\n",
    "print(x_test.shape)\n",
    "print(y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x[0]:[-1.14955978 -0.15830138 -0.7183361   0.45194578 -1.03505509  0.48853108\n",
      "  0.6231926   0.06563943 -0.4220751  -0.9940847  -0.42754369  0.94395616\n",
      " -1.17473827]\n",
      "\n",
      "X shape: (13,)\n"
     ]
    }
   ],
   "source": [
    "print(f\"x[0]:{x_train[0]}\\n\\nX shape: {x_train[0].shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Set up the neural network layers\n",
    "tf.random.set_seed(5)\n",
    "model=tf.keras.Sequential([\n",
    "    tf.keras.layers.Dense(26,activation='relu'),\n",
    "    tf.keras.layers.Dense(4,activation='softmax'),\n",
    "    \n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Compile the model\n",
    "model.compile(optimizer='adam',\n",
    "    loss=tf.keras.losses.SparseCategoricalCrossentropy(),\n",
    "    metrics=['accuracy'],)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 1.5282 - accuracy: 0.3145\n",
      "Epoch 2/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 1.4315 - accuracy: 0.3790\n",
      "Epoch 3/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 1.3419 - accuracy: 0.4194\n",
      "Epoch 4/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 1.2579 - accuracy: 0.5000\n",
      "Epoch 5/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 1.1817 - accuracy: 0.5161\n",
      "Epoch 6/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 1.1088 - accuracy: 0.5484\n",
      "Epoch 7/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 1.0415 - accuracy: 0.5645\n",
      "Epoch 8/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.9773 - accuracy: 0.6048\n",
      "Epoch 9/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.9186 - accuracy: 0.6210\n",
      "Epoch 10/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.8644 - accuracy: 0.6694\n",
      "Epoch 11/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.8120 - accuracy: 0.6935\n",
      "Epoch 12/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.7632 - accuracy: 0.7581\n",
      "Epoch 13/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.7189 - accuracy: 0.7823\n",
      "Epoch 14/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.6763 - accuracy: 0.7903\n",
      "Epoch 15/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.6378 - accuracy: 0.8226\n",
      "Epoch 16/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.5994 - accuracy: 0.8387\n",
      "Epoch 17/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.5654 - accuracy: 0.8548\n",
      "Epoch 18/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.5327 - accuracy: 0.8710\n",
      "Epoch 19/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.5028 - accuracy: 0.8710\n",
      "Epoch 20/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.4742 - accuracy: 0.8710\n",
      "Epoch 21/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.4490 - accuracy: 0.8952\n",
      "Epoch 22/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.4234 - accuracy: 0.9274\n",
      "Epoch 23/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.4023 - accuracy: 0.9355\n",
      "Epoch 24/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.3812 - accuracy: 0.9435\n",
      "Epoch 25/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.3621 - accuracy: 0.9435\n",
      "Epoch 26/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.3439 - accuracy: 0.9435\n",
      "Epoch 27/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.3271 - accuracy: 0.9516\n",
      "Epoch 28/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.3113 - accuracy: 0.9597\n",
      "Epoch 29/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.2970 - accuracy: 0.9597\n",
      "Epoch 30/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.2833 - accuracy: 0.9597\n",
      "Epoch 31/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.2709 - accuracy: 0.9597\n",
      "Epoch 32/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.2587 - accuracy: 0.9677\n",
      "Epoch 33/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.2481 - accuracy: 0.9677\n",
      "Epoch 34/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.2374 - accuracy: 0.9677\n",
      "Epoch 35/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.2280 - accuracy: 0.9677\n",
      "Epoch 36/60\n",
      "4/4 [==============================] - 0s 999us/step - loss: 0.2187 - accuracy: 0.9758\n",
      "Epoch 37/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.2102 - accuracy: 0.9758\n",
      "Epoch 38/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.2023 - accuracy: 0.9758\n",
      "Epoch 39/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1944 - accuracy: 0.9758\n",
      "Epoch 40/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1874 - accuracy: 0.9758\n",
      "Epoch 41/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1803 - accuracy: 0.9758\n",
      "Epoch 42/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1740 - accuracy: 0.9839\n",
      "Epoch 43/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1679 - accuracy: 0.9839\n",
      "Epoch 44/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1622 - accuracy: 0.9839\n",
      "Epoch 45/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1567 - accuracy: 0.9839\n",
      "Epoch 46/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.1512 - accuracy: 0.9839\n",
      "Epoch 47/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.1462 - accuracy: 0.9839\n",
      "Epoch 48/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1415 - accuracy: 0.9839\n",
      "Epoch 49/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1370 - accuracy: 0.9839\n",
      "Epoch 50/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1327 - accuracy: 0.9839\n",
      "Epoch 51/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1284 - accuracy: 0.9839\n",
      "Epoch 52/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1246 - accuracy: 0.9919\n",
      "Epoch 53/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.1207 - accuracy: 0.9919\n",
      "Epoch 54/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.1170 - accuracy: 0.9919\n",
      "Epoch 55/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1136 - accuracy: 0.9919\n",
      "Epoch 56/60\n",
      "4/4 [==============================] - ETA: 0s - loss: 0.1105 - accuracy: 1.00 - 0s 2ms/step - loss: 0.1105 - accuracy: 0.9919\n",
      "Epoch 57/60\n",
      "4/4 [==============================] - 0s 1ms/step - loss: 0.1073 - accuracy: 0.9919\n",
      "Epoch 58/60\n",
      "4/4 [==============================] - 0s 5ms/step - loss: 0.1044 - accuracy: 0.9919\n",
      "Epoch 59/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.1015 - accuracy: 0.9919\n",
      "Epoch 60/60\n",
      "4/4 [==============================] - 0s 2ms/step - loss: 0.0987 - accuracy: 0.9919\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x2402211a448>"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Train the model\n",
    "model.fit(x_train,y_train,epochs=60)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2/2 - 0s - loss: 0.1187 - accuracy: 1.0000\n",
      "\n",
      "Test accuracy: 100.0%\n"
     ]
    }
   ],
   "source": [
    "#Evaluate accuracy\n",
    "test_loss, test_acc = model.evaluate(x_test,y_test, verbose=2)\n",
    "\n",
    "print(f\"\\nTest accuracy: {round(test_acc*100,2)}%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.06018296, 0.02503642, 0.07485989, 0.83992076],\n",
       "       [0.10608073, 0.04703555, 0.6851922 , 0.16169152],\n",
       "       [0.02599302, 0.8836735 , 0.07199614, 0.01833734],\n",
       "       [0.00440406, 0.00674264, 0.9877158 , 0.00113754],\n",
       "       [0.2391886 , 0.06673394, 0.36645392, 0.32762358],\n",
       "       [0.02172054, 0.6867773 , 0.28271043, 0.00879173],\n",
       "       [0.01634306, 0.00849863, 0.09856146, 0.87659687],\n",
       "       [0.05586521, 0.7735848 , 0.1476263 , 0.02292364],\n",
       "       [0.01273871, 0.95428264, 0.02626936, 0.00670945],\n",
       "       [0.01668092, 0.97742647, 0.00294879, 0.00294387]], dtype=float32)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Make predictions\n",
    "predictions =model.predict(x_test)\n",
    "predictions[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "40\n"
     ]
    }
   ],
   "source": [
    "#In which Test set instance did the model make a wrong prediction?\n",
    "for i in range(len(x_test)):\n",
    "    if (predictions[i].argmax()!=y_test[i]):\n",
    "        wrong=i\n",
    "print(wrong)        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(wrong-1,wrong+3):\n",
    "    if (predictions[i].argmax()==y_test[i]):\n",
    "        color='green'\n",
    "    else:\n",
    "        color='red'\n",
    "    \n",
    "    fig = plt.figure()\n",
    "    ax = fig.add_subplot(111)\n",
    "    plt.ylabel('Probablity')\n",
    "    plt.xlabel(f'{str(predictions[i].max())[:4]}% is {predictions[i].argmax()} (True class is {y_test[i]})')\n",
    "    plt.xticks(range(4),list(range(4)))\n",
    "\n",
    "    ax.bar([0,1,2,3], predictions[i],  color=color, label='Bid')\n",
    "    ;"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X1 dimension is =(1,13)\n",
      "W1 dimension is =(13, 26)\n",
      "b1 dimension is =(26,)\n",
      "W2 dimension is =(26, 4)\n",
      "b2 dimension is =(4,)\n"
     ]
    }
   ],
   "source": [
    "#parameters and their shapes\n",
    "w1=model.weights[0].numpy()\n",
    "b1=model.weights[1].numpy()\n",
    "w2=model.weights[2].numpy()\n",
    "b2=model.weights[3].numpy()\n",
    "print(f\"X{1} dimension is =(1,13)\")\n",
    "print(f\"W{1} dimension is ={w1.shape}\\nb{1} dimension is ={b1.shape}\")\n",
    "print(f\"W{2} dimension is ={w2.shape}\\nb{2} dimension is ={b2.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "a2:         [0.06018296 0.02503642 0.07485989 0.83992074] \n",
      "\n",
      "TensorFlow: [0.06018296 0.02503642 0.07485989 0.83992076]\n"
     ]
    }
   ],
   "source": [
    "\n",
    "z1=np.matmul(x_test[0],w1)+b1\n",
    "a1=tf.keras.activations.relu(z1).numpy()\n",
    "z2=np.matmul(a1,w2)+b2\n",
    "a2=np.exp(z2)/np.sum(np.exp(z2))\n",
    "\n",
    "print(f\"a2:         {a2} \\n\\nTensorFlow: {predictions[0]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_2 (Dense)              (None, 26)                364       \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 4)                 108       \n",
      "=================================================================\n",
      "Total params: 472\n",
      "Trainable params: 472\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}