svm_classical_multiclass.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  _*SVM with a classical RBF kernel: multiclass classifier extension*_\n",
    "\n",
    "A multiclass extension works in conjunction with an underlying binary (two class) classifier to provide multiclass classification.\n",
    "\n",
    "Currently three different multiclass extensions are supported:\n",
    "\n",
    "* OneAgainstRest\n",
    "* AllPairs\n",
    "* ErrorCorrectingCode\n",
    "\n",
    "These use different techniques to group the data with binary classification to achieve the final multiclass classification."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datasets import *\n",
    "from qiskit_aqua.utils import split_dataset_to_data_and_labels\n",
    "from qiskit_aqua.input import SVMInput\n",
    "from qiskit_aqua import run_algorithm\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we choose the `Wine` dataset which has 3 classes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEICAYAAABbOlNNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3X24XGV57/Hvz5CQGNskQA6EhABRKiJS4GypBzwVeQnUVEIVMVprVGhqqyVHW2o4nOouisbaloZTrypFBGsrxFgxnOiJQABfUTYQgWADIehJwgYiENSQQAj3+WM9E9bszOyZvWftef19rmuumXnWWrPuWTtZ9zwv61mKCMzMzEpe0uoAzMysvTgxmJlZGScGMzMr48RgZmZlnBjMzKyME4OZmZVxYrAxISkkvSK9/pykv251TJVIOlnS5ibv87B0fPYp6PN+LWlOEZ9VYz/9kr481vux1nNi6FKSfiZpRzppPCbpakkvyy0/Q9J3JP1K0lZJt0k6a8hnnJxOYB9pJJaIeH9EfLyRz+gVki6S9K0hZQ9WKVsAEBEvi4iNzYyzlvTv7RPdsp9e48TQ3d4cES8Djgf6gP8FIOkc4KvAl4BZwIHAR4E3D9l+IfAk8O5mBdyIon6Bt9h3gBMljQOQNAMYDxw3pOwVaV2zwjkx9ICI2AJ8CzhakoB/AD4eEVdGxNMR8UJE3BYRf1zaRtJk4BzgA8ARkvqG24ekCyUNSnpE0vuGLNvzq67UdCPpryQ9nrY5W9KbJD0g6UlJ/7Pe75ZqRh+RdA+wXdI+kg6W9LVUE3pY0gW59SeleJ6SdD/w2iGft6cJbGjs6f18SWsl/VLSQ5LOTOVTJH0hfZ8tkj6RO5GPk/R3kn4haSMwb5ivdAdZIjg2vf/vwC3A+iFlD0XEI0NjTvF+VtKqVBv8kaSX5+I/UtKN6Tivl3TuMMf28FST/JWkG4EDhiz/qqRHJT2dap+vTuWLgD8E/irVWG9I5UvSMfuVpPsl/UHus16R9vV0Ok7X1Yq52n6sABHhRxc+gJ8Bp6XXhwDrgI8DRwIBHF5j+z8CBoFxwA3A/x5m3TOBx4CjgcnAv6d9vCItvxr4RHp9MvA8WQ1lPPDHwNa0zW8ArwZ21IpvyPdcm77jJLIfO3emz58AzAE2Amek9ZcC3wX2S9vcB2zOfd6euCvEfgLwNHB62s9M4Mi07OvA59P3/y/Aj4E/ScveD/xn2t9+ZCf6APap8p1uAT6UXv8T8D7g0iFlV1WKOcX7RIp1H+DfgGvTssnAJuC9adlxwC+Ao6rE8UOyHxH7Ar8L/Ar4cm75+9LfbF/gH4G1lY5bruxtwMHp2L0d2A7MSMu+Alyclk0EXl9PzJX240fjD9cYutv1krYB3wNuAz4J7J+WDdbYdiFwXUTsJjtpL5A0vsq65wJfjIj7ImI70F/js3cBl0bELuBasl+iyyLiVxGxDrgf+O0an5F3eURsiogdZDWA6RFxSUQ8F1nb+78AC3KxXhoRT0bEJuDyEeznPLIT8o2R1bK2RMR/SjoQeBPwPyJie0Q8Dlw2ZJ//mGJ8EvhUjf3cRnYihqx28N30yJfdNsz2X4+IH0fE82SJoVTT+H3gZxHxxYh4PiLuBr5GdsIuI2k22bH864h4NiK+Q/YDYY+IuCr9zZ4l+5v/tqQp1YKKiK9GxCPp2F0HPEiWwCD7N3EocHBE7IyI7400ZiuOE0N3OzsipkbEoRHxZ+nE+URaNqPaRpIOAd5IdlIB+AbZr7hqTSAHk/2qK/l5jbieSAkHstoBZDUOcmUvo375fR8KHCxpW+kB/E+yfpTRxJp3CPBQhfJDyWo/g7l9fp6s5jCafX4HeL2k/ciS3IPAD8j6HvYjq5kN17/waO71M7x4LA8FfmfIsflD4KAKn3Ew8FRK9HvFnZrHlqamoV+S1dxgSHNTnqR3p2a40r6Pzq3/V4CAH0tal2uOHEnMVpBu6KyzkVlPdpJ6K/B3Vdb5I7IfDTdkXRJAlhgWAtdXWH+Q7KRZMruQSOuXnyJ4E/BwRBxRZd1SrOvS+6GxPgO8NPf+IKA0nHUT8HL2tgl4Fjgg/Uqvts+SWsfnh8AUsma27wNExC8lPZLKHomIh2t8RiWbgNsi4vQ61h0EpkmanEsOs3nxWL8TmA+cRpYUpgBPkZ3cofxvgqRDyWpupwI/jIjdktaW1o+IR9N3Q9LrgZskfaeOmD099BhwjaHHREQAHwb+WtJ7Jf2mpJdIer2kK9JqC4G/IWuCKD3eCrxJ0v4VPnY58B5JR0l6KfCxsf8mVf0Y+FXqkJ6UftkeLanUybwcuEjSNEmzgD8fsv1a4J1puzOBN+SWfQF4r6RT0zGbKenIiBgEvg38fe54vlxSadvlwAWSZkmaBiwZ7gukmt0A2d/pu7lF30tlox2N9H+A35L0R5LGp8drJb2qQgw/TzH8jaQJ6WSdH7X2G2TJ8AmyRPrJIR/xGFn/TslkspP4VgBJ7yWrMZDevy39PSBLMAG8UEfMQ/djBXBi6EERsYKs8+99wCNk/7k+AXxD0uvIqu+fjYhHc4+VwAbgHRU+71tknY9r0jpriopV0h9KWld7zT2x7CZrlz4WeJiso/JKsl+0kCW8n6dl3wb+dchHLCY7AZaaLPbUkCLix2SdoJeRdULfRnasIBvSO4Gsf+QpYAUvNtf9C7Aa+AlwF/AfdXyV28iaor6XK/tuKhtVYoiIXwFzyfo+HiFrcvo0WedxJe8EfodsyPLHyIY3l3yJ7DhuIfvOtw/Z9gvAUan55/qIuB/4e7La0GPAa0i1oeS1wI8k/RpYCSyOiI11xFy2nxEcDhuGsh+QZmZmGdcYzMysjBODmZmVcWIwM7MyTgxmZlamI69jOOCAA+Kwww5rdRhmZh3lzjvv/EVETK+1XkcmhsMOO4yBgYFWh2Fm1lEk1XWlv5uSzMysjBODmZmVcWIwM7MyTgxmZlbGicHMzMo4MZiZWZlCEoOkq5Tdv/e+Kssl6XJJGyTdI+n43LKFkh5Mj4VFxGM97p7lcNnR0D81e75neasjMusoRdUYria77281vwcckR6LgH8GSHej+hjZ1L4nAB9L89Wbjc49y+GGC+DpTUBkzzdc4ORgNgKFJIZ0P9gnh1llPvClyNwOTJU0AzgDuDHdf/cp4EaGTzBmw7v5Eti1o7xs146s3Mzq0qw+hpmU3/N2cyqrVr4XSYskDUga2Lp165gFah3u6c0jKzezvXRM53NEXBERfRHRN316zak+rFdNmTWycjPbS7MSwxbKb4Y+K5VVKzcbnVM/CuMnlZeNn5SVm1ldmpUYVgLvTqOTXgc8nW6gvhqYm27MPo3s3q6rmxSTdaNjzoU3Xw5TDgGUPb/58qzczOpSyOyqkr4CnAwcIGkz2Uij8QAR8Tngm8CbyG4U/wzZDdWJiCclfRy4I33UJRExXCe2WW3HnOtEYNaAQhJDRLyjxvIAPlBl2VXAVUXEYWZmjeuYzmczM2sOJwYzMyvjxGBmZmWcGMzMrIwTg5mZlXFiMDOzMk4MZmZWxonBzMzKODGYmVkZJwYzMyvjxGBmZmWcGMzMrIwTg5mZlXFiMDOzMk4MZmZWppDEIOlMSeslbZC0pMLyyyStTY8HJG3LLdudW7ayiHjMzGz0Gr5Rj6RxwGeB04HNwB2SVkbE/aV1IuJDufX/HDgu9xE7IuLYRuMwM7NiFFFjOAHYEBEbI+I54Fpg/jDrvwP4SgH7NTOzMVBEYpgJbMq935zK9iLpUOBwYE2ueKKkAUm3Szq7gHjMzKwBhdzzeQQWACsiYneu7NCI2CJpDrBG0r0R8dDQDSUtAhYBzJ49uznRmpn1oCJqDFuAQ3LvZ6WyShYwpBkpIrak543ArZT3P+TXuyIi+iKib/r06Y3GbGZmVRSRGO4AjpB0uKQJZCf/vUYXSToSmAb8MFc2TdK+6fUBwEnA/UO3NTOz5mm4KSkinpf0QWA1MA64KiLWSboEGIiIUpJYAFwbEZHb/FXA5yW9QJakluZHM5mZWfOp/DzdGfr6+mJgYKDVYZiZdRRJd0ZEX631fOWztcY9y+Gyo6F/avZ8z/JWR2RmSbNHJZllSeCGC2DXjuz905uy9wDHnNu6uMwMcI3BWuHmS15MCiW7dmTlZtZyTgzWfE9vHlm5mTWVE4M135RZIys3s6ZyYrDmO/WjMH5Sedn4SVm5mbWcE4M13zHnwpsvhymHAMqe33x593Y8ewSWdRiPSrLWOObc7k0EeR6BZR3INQazseQRWNaBnBjMxpJHYFkHcmIwG0ujHYHlfglrIScGs7FUaQQWwHPbq5/sS/0ST28C4sV+CScHaxInBrOiVPqVXxqBNWm/8nV3PFn9ZO9+CWsxJwazIgz3K/+Yc2HC5L23qXayd7+EtZgTg7WvTmpnr/UrfyQne18Zbi3mxGDtqZ3a2etJULVO/CM52fvKcGuxQhKDpDMlrZe0QdKSCsvfI2mrpLXpcX5u2UJJD6bHwiLisS7QLu3s9SaoWif+kZzse+3KcGs7DV/5LGkc8FngdGAzcIeklRVu0XldRHxwyLb7AR8D+oAA7kzbPtVoXNbh2qWdfbgElT9Rn/rR8iucofzEX1r35kuy7zBlVras2sm+V64Mt7ZUxJQYJwAbImIjgKRrgflAPfduPgO4MSKeTNveCJwJfKWAuLrXPcuHP8HUWt4JpsxKv9IrlDdTvQmqnhO/T/bWIYpIDDOB/P/gzcDvVFjvrZJ+F3gA+FBEbKqy7cxKO5G0CFgEMHv27ALC7lC15t7plrl5av0Cb5ZaCaoXkrT1nGZ1Pt8AHBYRxwA3AteM9AMi4oqI6IuIvunTpxceYMeo1fbeLm3zjWqXdvbh+gZq9T+0Uwe62QgUUWPYAhySez8rle0REU/k3l4J/G1u25OHbHtrATF1r1pNG+3SNl+Edmh6Ga6J6LKjh+9/qLd/wqzNFJEY7gCOkHQ42Yl+AfDO/AqSZkTEYHp7FvDT9Ho18ElJ09L7ucBFBcTUvWo1bbRL23w3qZageilJW09puCkpIp4HPkh2kv8psDwi1km6RNJZabULJK2T9BPgAuA9adsngY+TJZc7gEtKHdFWRa1hjx4D3zy1hqj6QjXrUIqIVscwYn19fTEwMNDqMFrHHZ7tYWhHP2RJuNQXUmu5WZNJujMi+mqu58Rg1gAnaesgTgxmZlam3sTguZLMzKyME4OZmZVxYjAzszJODFZbJ90XwcwaVsQFbtbNumXuJTOrm2sMNrxumXvJzOrmxGDDG8tpHdxEZdaWnBh6WT0n5rGa1sEzj5q1LSeGXlXviXms5l5yE5VZ23Ji6FX1npjH6r4InnnUrG15VFKvGsmJeSzui+Dpwc3almsMvarVU0J7enCztuXE0KtafWJul1t3mtleCmlKknQmsAwYB1wZEUuHLP8wcD7wPLAVeF9E/Dwt2w3cm1b9fxFxFjb2hrtlZTNjcCIwazsNJwZJ44DPAqcDm4E7JK2MiPtzq90N9EXEM5L+lOyez29Py3ZExLGNxmGj4BOzmVVQRFPSCcCGiNgYEc8B1wLz8ytExC0R8Ux6ezvgHkYzszZVRGKYCeSHl2xOZdWcB3wr936ipAFJt0s6u9pGkhal9Qa2bt3aWMRmZlZVU4erSnoX0Ae8IVd8aERskTQHWCPp3oh4aOi2EXEFcAVkd3BrSsBmZj2oiBrDFuCQ3PtZqayMpNOAi4GzIuLZUnlEbEnPG4FbgeMKiMnMzEapiMRwB3CEpMMlTQAWACvzK0g6Dvg8WVJ4PFc+TdK+6fUBwElAvtPaOpUnyDPrWA03JUXE85I+CKwmG656VUSsk3QJMBARK4HPAC8DvioJXhyW+irg85JeIEtSS4eMZrJONNp7ONyzvLXDZ80MAEV0XnN9X19fDAwMtDoMq+ayo6tMd3EIfOi+ytsMTSaQXXDni97MCiPpzojoq7Wer3y24o1mgjzPtmrWNpwYrHijmYfJs62atQ0nBiveaOZhavWkfma2hxODFW80E+S1elI/M9vD92OwsTHSeZjaYVK/TudRXVYQJwZrH57Ub/RGO0TYrAI3JZl1A4/qsgI5MZh1A4/qsgI5MZh1gw4e1bVq4yrmrpjLMdccw9wVc1m1cVWrQ+p57mMwrr97C59ZvZ5Htu3g4KmTuPCMV3L2ccPNnG6jNlYdxKd+tPKV420+qmvVxlX0/6Cfnbt3AjC4fZD+H/QDMG/OvBZG1ttcY+hx19+9hYv+4162bNtBAFu27eCi/7iX6+/ea4Jca1Spg/jpTUC82EFcxASDHXoP7WV3LduTFEp27t7JsruWtSgiA9cYet5nVq9nx67dZWU7du3mM6vXu9ZQtOE6iIs4gXfgqK5Htz86onJrDtcYetwj23aMqNwa4A7ivRw0+aARlVtzODH0uIOnThpRuTWggzuIx8ri4xczcdzEsrKJ4yay+PjFLYrIwImhbVx/9xZOWrqGw5es4qSla5rWxn/hGa9k0vhxZWWTxo/jwjNeOarPa9X36Aie9mMv8+bMo//EfmZMnoEQMybPoP/Efnc8t1gh92OQdCawjOxGPVdGxNIhy/cFvgT8V+AJ4O0R8bO07CLgPGA3cEFErK61v267H0OpAzjf1j9p/Dg+9ZbXNKWdv6hRSa3+Hh3B01ZYC9V7P4aGE4OkccADwOnAZrJbfb4jfyc2SX8GHBMR75e0APiDiHi7pKOArwAnAAcDNwG/FRG7h+4nr9sSw0lL17ClQpv+zKmT+P6SU1oQ0eh0y/cw61bNvFHPCcCGiNgYEc8B1wLzh6wzH7gmvV4BnKrsHp/zgWsj4tmIeBjYkD6vp3RLB3C3fA+zXldEYpgJ5O/juDmVVVwnIp4Hngb2r3NbACQtkjQgaWDr1q0FhN0+uqUDuFu+h1mv65jO54i4IiL6IqJv+vTprQ6nUEV3ALdKt3wPs15XxAVuW4BDcu9npbJK62yWtA8whawTup5tu16pY7bTp6Wo9j0g63/o5O9m1kuK6Hzeh6zz+VSyk/odwDsjYl1unQ8Ar8l1Pr8lIs6V9Grg33mx8/lm4Ihe63weqU6a28gjlczaR9M6n1OfwQeB1cBPgeURsU7SJZLOSqt9Adhf0gbgw8CStO06YDlwP/B/gQ/USgo94Z7lcNnR0D81e87NpdNpcxsNN+WGmbWnQuZKiohvAt8cUvbR3OudwNuqbHspcGkRcXSFGnfi6rS5jYoaqdRJtSSzTtcxnc89o8aduDptSGgRI5U6rZZk1umcGNpNjYnWOm1IaBEjldwcZdZcTgztpsZEa502JPTs42byqbe8hplTJyGyq6BH2vHcabUks07n+zG0mxp34urEoa1nHzezofgOnjqp4lQb7VpLMut0TgztpjSh2jATrTV6ou00F57xyopDXtu1lmQjs2rjKpbdtYxHtz/KQZMPYvHxiz27aos5MbSjDrwT11jqxFqS1aeZ93x2AqpfIdNuN1uvX+Bm1i3mrpjL4PbBvcpnTJ7Bt8/5dmH7GZqAILshUK/d+6GZs6taB/ONdayVmnXP52V3LStLCgA7d+9k2V3LCt1Pt3Bi6GG+PsBarVn3fG5WAuoWTgw9zNcHWKs1657PzUpA3cKJoYf5+gBrtWbd87lZCahbeFTSGGjmvD6N7GvKpPFs27Frr3JfH2DNNG/OvEISwXCjjkrPHpVUHyeGgg2dZrrUbg8Unhwa2df1d29h+3PP71U+/iXy9QHWceoZ9lpUAuoFbkoqWDPb7RvZ12dWr2fX7r2HKr9s4j6+PsA6jkcdFcs1hoKNtN2+kaagRvoIqq2z7Zm9m5bMWqmeC9O6edRRKy7Mc42hYCOZ/bTR4aKNzLTaabO0Wm8qNRENbh8kiD1NRKs2ripbr1tHHdX7/YvWUGKQtJ+kGyU9mJ6nVVjnWEk/lLRO0j2S3p5bdrWkhyWtTY9jG4mnHYxk9tNqTUF/sfwndSWHRmZa7bRZWq031dtE1K2jjlrVRNZoU9IS4OaIWCppSXr/kSHrPAO8OyIelHQwcKek1RGxLS2/MCJWNBhH2xjJvD7VmnN2R9TVidzIHEKef8g6Qb1NRN066qhVTWSNJob5wMnp9TXArQxJDBHxQO71I5IeB6YD2+hS9c5+Wm06aaj/dp2NzLTaa7O0Wuc5aPJBFedSqtRE1I2jjkby/YvUaB/DgRFRivpR4MDhVpZ0AjABeChXfGlqYrpM0r7DbLtI0oCkga1btzYYdnuo1JyT5wvNrNd1axNRvVr1/WvWGCTdBFRKTxfn30RESKo6VaukGcC/Agsj4oVUfBFZQpkAXEFW27ik0vYRcUVah76+vs6bEraC0q/1v1j+E3ZXmOXWHcHW67q1iaherfr+DU27LWk9cHJEDKYT/60RsVfvpaTfJGtm+mS1/gRJJwN/GRG/X2u/RUy73cyrk+uJpdKNaEZ6C0wzs+E0a9rtlcDC9Hoh8I0KgUwAvg58aWhSSMkESQLOBu5rMJ66tNusokXcF9nMrCiN1hj2B5YDs4GfA+dGxJOS+oD3R8T5kt4FfBFYl9v0PRGxVtIaso5oAWvTNr+utd9GawwnLV1TsdN35tRJfH/JKaP+XDOzdlZvjaGhUUkR8QRwaoXyAeD89PrLwJerbN+Ss3ArZhVtp6YrM7Ph9OSVz82+6rfdmq7MWmHVxlXMXTGXY645hrkr5o751bs2ej2ZGJp91a9viGO9rlVTO9jo9GRiaHZnr2+IY73Os592lp6dXbWZV/1Wu8LZ1ylYr+jm2U+7UU/WGJrNE9ZZr+vW2U+7lRNDE/g6Bet1vT61Rafp2aakZvOEddbLen1qi07jxGBmTdGNs592KzclmZlZmZ6rMXT7Fcjd/v3MbOz1VGIYOotp6QpkGP5OaZ1iuO8HvlubmdWnpxLDcFcgd8NJstr3+5sb1rFz1wtdmxDNSlZtXOUO7gL0VB9Dt1+BXO17PPXMLk/JYV3P024Up6cSQ7Mnz8u7/u4tnLR0DYcvWcVJS9eMyQR6I/0e3ZIQzcDTbhSppxJDq65AbtbsqtW+39RJ4yuu7yk5rJt42o3iNJQYJO0n6UZJD6bnaVXW2y1pbXqszJUfLulHkjZIui7d7W3MtOoK5GbNrlrt+/Wf9eqWJcSxriWZlXjajeI02vm8BLg5IpZKWpLef6TCejsi4tgK5Z8GLouIayV9DjgP+OcGYxpWK65AbmbfxnDfr5mjkrp9BJi1n8XHL6b/B/1lzUmedmN0Gk0M84GT0+trgFupnBj2ku7zfArwztz2/YxxYmiFdphdtdkJsdtHgFlzjGSUkafdKE6jieHAiBhMrx8FDqyy3kRJA8DzwNKIuB7YH9gWEc+ndTYDVc8YkhYBiwBmz57dYNjNdeEZryz79QzdP7tqt48As7FXGmVUqgGURhkBwyYHJ4LG1exjkHSTpPsqPObn14uIAKLKxxyabkD9TuAfJb18pIFGxBUR0RcRfdOnTx/p5i3Vi7OrtnIEmHWHdh5l1O23Ka1ZY4iI06otk/SYpBkRMShpBvB4lc/Ykp43SroVOA74GjBV0j6p1jAL6NreyV6bXbUXa0lWrHYdZTSamkynaXS46kpgYXq9EPjG0BUkTZO0b3p9AHAScH+qYdwCnDPc9taZerGWZMVq11FG7VyTKUqjfQxLgeWSzgN+DpwLIKkPeH9EnA+8Cvi8pBfIEtHSiLg/bf8R4FpJnwDuBr7QYDxN4Ynq6tNrtSQr1khGGTVzKox2rckUqaHEEBFPAKdWKB8Azk+vfwC8psr2G4ETGomh2TwM06w56h1l1OymnYMmH8Tg9sGK5d1CWYtOZ+nr64uBgYGW7PukpWsqDj2dOXUS319ySgsiMuttc1fMrXiinjF5Bt8+59uF729oIoKsJtN/Yn/b9zFIujMNBBpWT82uWgQPwzSrT7Oad6o14VRKFkXoheslnBhGqB0uVjNrd81s3qnWtFOKYyxO2N1+vURPTaJXhFZNxGfWSZo5cme4KS+6aaRQM7nGMEKlDmaPSjKrbixH7lRqohppHDY8J4ZR8DBMs+GN1cidak1UU/edyrZntxWyP98Fzk1JZjYGFh+/mInjJpaVFTHTabUmqogoZH++C1zGicHMCjdvzjz6T+xnxuQZCDFj8oyKwzlHOudQtaahXz73y7r2V0svXNVcDzclmdmYqDVyZzQjl4ZroipipFAvXNVcD9cYzKypSrWEJd9dMuJf52PVRFXSrvMzNZsTg5k1Tb4Nv5rhfp3X20Q1WmOdeDqFm5LMelizR+BUasMfqtav87G8uKwXrmquhxODWY9qxX0FarXVt8Ov826/qrkebkoy61GtGIEzXG2g6GYhGz0nBrMe1YoROJXa8AGm7ju1J5ts2lVDiUHSfpJulPRgep5WYZ03Slqbe+yUdHZadrWkh3PLjm0kHjOrXytG4JQ6j6dMmFJWvu3ZbT15IVm7arTGsAS4OSKOAG5O78tExC0RcWxEHAucAjwD5CdJv7C0PCLWNhiPmdWpVSNw5s2Zx0vHv3Sv8l68kKxdNdr5PB84Ob2+BriV7Had1ZwDfCsinmlwv2bWoFaOwPGFZO2t0cRwYESUBiQ/ChxYY/0FwD8MKbtU0kdJNY6IeLbShpIWAYsAZs+ePfqIzWyPVo3A6YXbY3aymk1Jkm6SdF+Fx/z8epHdI7TqfUIlzSC79/PqXPFFwJHAa4H9GKa2ERFXRERfRPRNnz69Vthm1sZ8IVl7q1ljiIjTqi2T9JikGRExmE78jw/zUecCX4+IXbnPLv1keFbSF4G/rDNuM+tgvpCsvTXalLQSWAgsTc/fGGbdd5DVEPbIJRUBZwP3NRiPmXUIX0jWvhodlbQUOF3Sg8Bp6T2S+iRdWVpJ0mHAIcBtQ7b/N0n3AvcCBwCfaDAeMzNrUEM1hoh4Aji1QvkAcH7u/c+AvW55FhGnNLJ/MzMrnq98NjOzMk4MZmZWxonBzMzKODGYmVkZJwYzMyvjxGBmZmWcGMzMrIwTg5mZlXFiMDOzMk4MZmZWxonBzMzKODGYmVkZJwYzMyvjxGBmZmWcGMzMrExDiUHS2yStk/SCpL5h1jtT0npJGyQtyZUfLulHqfw6SRMaicfMzBoZ9NI0AAAHbElEQVTXaI3hPuAtwHeqrSBpHPBZ4PeAo4B3SDoqLf40cFlEvAJ4CjivwXjMutKqjauYu2Iux1xzDHNXzGXVxlWtDsm6WEOJISJ+GhHra6x2ArAhIjZGxHPAtcD8dJ/nU4AVab1ryO77bGY5qzauov8H/QxuHyQIBrcP0v+DficHGzPN6GOYCWzKvd+cyvYHtkXE80PKzSxn2V3L2Ll7Z1nZzt07WXbXshZFZN2u5j2fJd0EHFRh0cUR8Y3iQ6oaxyJgEcDs2bObtVuzlnt0+6MjKjdrVM3EEBGnNbiPLcAhufezUtkTwFRJ+6RaQ6m8WhxXAFcA9PX1RYMxmXWMgyYfxOD2wYrlZmOhGU1JdwBHpBFIE4AFwMqICOAW4Jy03kKgaTUQs06x+PjFTBw3saxs4riJLD5+cYsism7X6HDVP5C0GfhvwCpJq1P5wZK+CZBqAx8EVgM/BZZHxLr0ER8BPixpA1mfwxcaicesG82bM4/+E/uZMXkGQsyYPIP+E/uZN2deq0OzLqXsh3tn6evri4GBgVaHYWbWUSTdGRFVrzkr8ZXPZmZWxonBzMzKODGYmVkZJwYzMyvjxGBmZmWcGMzMrIwTg5mZlenI6xgkbQV+3qTdHQD8okn7Gq1OiBE6I85OiBE6I85OiBF6K85DI2J6rZU6MjE0k6SBei4IaaVOiBE6I85OiBE6I85OiBEcZyVuSjIzszJODGZmVsaJobYrWh1AHTohRuiMODshRuiMODshRnCce3Efg5mZlXGNwczMyjgxmJlZmZ5PDJL2k3SjpAfT87QK67xR0trcY6eks9OyqyU9nFt2bKviTOvtzsWyMld+uKQfSdog6bp0N72WxCnpWEk/lLRO0j2S3p5bNmbHU9KZktanY7CkwvJ907HZkI7VYbllF6Xy9ZLOKCqmUcT4YUn3p+N2s6RDc8sq/u1bFOd7JG3NxXN+btnC9O/jQUkLWxznZbkYH5C0LbesKcdT0lWSHpd0X5XlknR5+g73SDo+t2xsjmVE9PQD+FtgSXq9BPh0jfX3A54EXpreXw2c0y5xAr+uUr4cWJBefw7401bFCfwWcER6fTAwCEwdy+MJjAMeAuYAE4CfAEcNWefPgM+l1wuA69Lro9L6+wKHp88Z16IY35j7t/enpRiH+9u3KM73AP9UYdv9gI3peVp6Pa1VcQ5Z/8+Bq1pwPH8XOB64r8ryNwHfAgS8DvjRWB/Lnq8xAPOBa9Lra4Cza6x/DvCtiHhmTKPa20jj3EOSgFOAFaPZfoRqxhkRD0TEg+n1I8DjQM2rMRt0ArAhIjZGxHPAtSnWvHzsK4BT07GbD1wbEc9GxMPAhvR5TY8xIm7J/du7HZg1BnHUUs+xrOYM4MaIeDIingJuBM5skzjfAXxljGKpKiK+Q/Zjs5r5wJciczswVdIMxvBYOjHAgRExmF4/ChxYY/0F7P2P59JUxbtM0r6FR5ipN86JkgYk3V5q7iK7n/a2yO6/DbAZmNniOAGQdALZr7mHcsVjcTxnApty7ysdgz3rpGP1NNmxq2fbZsWYdx7ZL8mSSn/7sVBvnG9Nf8cVkg4Z4bZFqHtfqUnucGBNrrhZx7OWat9jzI7lPkV8SLuTdBNwUIVFF+ffRERIqjp+N2Xp1wCrc8UXkZ0AJ5CNM/4IcEkL4zw0IrZImgOskXQv2QmuMAUfz38FFkbEC6m4sOPZzSS9C+gD3pAr3utvHxEPVf6EMXcD8JWIeFbSn5DVxE5pUSz1WACsiIjdubJ2Op5N1ROJISJOq7ZM0mOSZkTEYDpRPT7MR50LfD0iduU+u/Tr+FlJXwT+spVxRsSW9LxR0q3AccDXyKqf+6RfwrOALa2MU9JvAquAi1P1uPTZhR3PIbYAh+TeVzoGpXU2S9oHmAI8Uee2zYoRSaeRJeE3RMSzpfIqf/uxOJHVjDMinsi9vZKs76m07clDtr218Ahf3Fe9f7cFwAfyBU08nrVU+x5jdizdlAQrgVJv/kLgG8Osu1cbZDr5ldrxzwYqjiwoQM04JU0rNb1IOgA4Cbg/sp6qW8j6R6pu38Q4JwBfJ2s3XTFk2VgdzzuAI5SNzppAdiIYOtIkH/s5wJp07FYCC5SNWjocOAL4cUFxjShGSccBnwfOiojHc+UV//ZjEGO9cc7IvT0L+Gl6vRqYm+KdBsylvAbe1DhTrEeSdd7+MFfWzONZy0rg3Wl00uuAp9MPqLE7ls3odW/nB1kb8s3Ag8BNwH6pvA+4MrfeYWQZ+iVDtl8D3Et2Avsy8LJWxQmcmGL5SXo+L7f9HLKT2Qbgq8C+LYzzXcAuYG3ucexYH0+y0R0PkP3quziVXUJ2kgWYmI7NhnSs5uS2vThttx74vTH891grxpuAx3LHbWWtv32L4vwUsC7FcwtwZG7b96VjvAF4byvjTO/7gaVDtmva8ST7sTmY/k9sJus7ej/w/rRcwGfTd7gX6BvrY+kpMczMrIybkszMrIwTg5mZlXFiMDOzMk4MZmZWxonBzMzKODGYmVkZJwYzMyvz/wHmBkF0tonGcwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "feature_dim = 2  # dimension of each data point\n",
    "sample_Total, training_input, test_input, class_labels = Wine(training_size=20,\n",
    "                                                              test_size=10, n=feature_dim, PLOT_DATA=True)\n",
    "\n",
    "temp = [test_input[k] for k in test_input]\n",
    "total_array = np.concatenate(temp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we setup an Aqua configuration dictionary to use the classical `SVM` algorithm and add a multiclass extension to classify the Wine data set, since it has 3 classes. We loop over the three extensions (modifying the params dictionary) to show the result with each."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "----- Using multiclass extension: 'OneAgainstRest' -----\n",
      "\n",
      "'testing_accuracy' : 1.0\n",
      "'test_success_ratio' : 1.0\n",
      "'predicted_labels' : [0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2]\n",
      "'predicted_classes' : ['A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'C', 'C', 'C', 'C', 'C', 'C', 'C', 'C', 'C', 'C']\n",
      "\n",
      "----- Using multiclass extension: 'AllPairs' -----\n",
      "\n",
      "'testing_accuracy' : 1.0\n",
      "'test_success_ratio' : 1.0\n",
      "'predicted_labels' : [0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2]\n",
      "'predicted_classes' : ['A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'C', 'C', 'C', 'C', 'C', 'C', 'C', 'C', 'C', 'C']\n",
      "\n",
      "----- Using multiclass extension: 'ErrorCorrectingCode' -----\n",
      "\n",
      "'testing_accuracy' : 1.0\n",
      "'test_success_ratio' : 1.0\n",
      "'predicted_labels' : [0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2]\n",
      "'predicted_classes' : ['A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'A', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'B', 'C', 'C', 'C', 'C', 'C', 'C', 'C', 'C', 'C', 'C']\n"
     ]
    }
   ],
   "source": [
    "aqua_dict = {\n",
    "    'problem': {'name': 'svm_classification'},\n",
    "    'algorithm': {\n",
    "        'name': 'SVM'\n",
    "    },\n",
    "    'multiclass_extension': {'name': 'OneAgainstRest'}\n",
    "}\n",
    "\n",
    "algo_input = SVMInput(training_input, test_input, total_array)\n",
    "\n",
    "extensions = [\n",
    "   {'name': 'OneAgainstRest'},\n",
    "   {'name': 'AllPairs'}, \n",
    "   {'name': 'ErrorCorrectingCode', 'code_size': 5}\n",
    "]\n",
    "\n",
    "for extension in extensions:\n",
    "    aqua_dict['multiclass_extension'] = extension\n",
    "    result = run_algorithm(aqua_dict, algo_input)\n",
    "    print(\"\\n----- Using multiclass extension: '{}' -----\\n\".format(extension['name']))\n",
    "    for k,v in result.items():\n",
    "        print(\"'{}' : {}\".format(k, v))\n"
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}