MultiClassExample.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/200\n",
      "23/23 - 0s - loss: 1.6717 - accuracy: 0.2768\n",
      "Epoch 2/200\n",
      "23/23 - 0s - loss: 1.2780 - accuracy: 0.2768\n",
      "Epoch 3/200\n",
      "23/23 - 0s - loss: 1.2164 - accuracy: 0.3750\n",
      "Epoch 4/200\n",
      "23/23 - 0s - loss: 1.3239 - accuracy: 0.2857\n",
      "Epoch 5/200\n",
      "23/23 - 0s - loss: 1.1382 - accuracy: 0.3929\n",
      "Epoch 6/200\n",
      "23/23 - 0s - loss: 1.0720 - accuracy: 0.3839\n",
      "Epoch 7/200\n",
      "23/23 - 0s - loss: 1.3171 - accuracy: 0.3036\n",
      "Epoch 8/200\n",
      "23/23 - 0s - loss: 1.1219 - accuracy: 0.4018\n",
      "Epoch 9/200\n",
      "23/23 - 0s - loss: 1.1083 - accuracy: 0.5268\n",
      "Epoch 10/200\n",
      "23/23 - 0s - loss: 1.0012 - accuracy: 0.5357\n",
      "Epoch 11/200\n",
      "23/23 - 0s - loss: 0.9251 - accuracy: 0.5804\n",
      "Epoch 12/200\n",
      "23/23 - 0s - loss: 0.9283 - accuracy: 0.5625\n",
      "Epoch 13/200\n",
      "23/23 - 0s - loss: 1.0167 - accuracy: 0.5625\n",
      "Epoch 14/200\n",
      "23/23 - 0s - loss: 0.8663 - accuracy: 0.6429\n",
      "Epoch 15/200\n",
      "23/23 - 0s - loss: 0.8743 - accuracy: 0.6607\n",
      "Epoch 16/200\n",
      "23/23 - 0s - loss: 0.9588 - accuracy: 0.6518\n",
      "Epoch 17/200\n",
      "23/23 - 0s - loss: 0.8249 - accuracy: 0.6518\n",
      "Epoch 18/200\n",
      "23/23 - 0s - loss: 0.8622 - accuracy: 0.6696\n",
      "Epoch 19/200\n",
      "23/23 - 0s - loss: 0.8060 - accuracy: 0.6786\n",
      "Epoch 20/200\n",
      "23/23 - 0s - loss: 0.7088 - accuracy: 0.6964\n",
      "Epoch 21/200\n",
      "23/23 - 0s - loss: 0.6897 - accuracy: 0.7411\n",
      "Epoch 22/200\n",
      "23/23 - 0s - loss: 0.8003 - accuracy: 0.6339\n",
      "Epoch 23/200\n",
      "23/23 - 0s - loss: 0.7729 - accuracy: 0.6518\n",
      "Epoch 24/200\n",
      "23/23 - 0s - loss: 0.6475 - accuracy: 0.6964\n",
      "Epoch 25/200\n",
      "23/23 - 0s - loss: 0.6557 - accuracy: 0.6964\n",
      "Epoch 26/200\n",
      "23/23 - 0s - loss: 0.6486 - accuracy: 0.6786\n",
      "Epoch 27/200\n",
      "23/23 - 0s - loss: 0.5978 - accuracy: 0.7589\n",
      "Epoch 28/200\n",
      "23/23 - 0s - loss: 0.6365 - accuracy: 0.6696\n",
      "Epoch 29/200\n",
      "23/23 - 0s - loss: 0.5830 - accuracy: 0.6875\n",
      "Epoch 30/200\n",
      "23/23 - 0s - loss: 0.6555 - accuracy: 0.7143\n",
      "Epoch 31/200\n",
      "23/23 - 0s - loss: 0.5969 - accuracy: 0.7054\n",
      "Epoch 32/200\n",
      "23/23 - 0s - loss: 0.5799 - accuracy: 0.7054\n",
      "Epoch 33/200\n",
      "23/23 - 0s - loss: 0.5916 - accuracy: 0.6875\n",
      "Epoch 34/200\n",
      "23/23 - 0s - loss: 0.5745 - accuracy: 0.6786\n",
      "Epoch 35/200\n",
      "23/23 - 0s - loss: 0.5643 - accuracy: 0.7143\n",
      "Epoch 36/200\n",
      "23/23 - 0s - loss: 0.5508 - accuracy: 0.6696\n",
      "Epoch 37/200\n",
      "23/23 - 0s - loss: 0.5495 - accuracy: 0.6786\n",
      "Epoch 38/200\n",
      "23/23 - 0s - loss: 0.4861 - accuracy: 0.8393\n",
      "Epoch 39/200\n",
      "23/23 - 0s - loss: 0.5252 - accuracy: 0.7054\n",
      "Epoch 40/200\n",
      "23/23 - 0s - loss: 0.5072 - accuracy: 0.7411\n",
      "Epoch 41/200\n",
      "23/23 - 0s - loss: 0.4839 - accuracy: 0.7946\n",
      "Epoch 42/200\n",
      "23/23 - 0s - loss: 0.4883 - accuracy: 0.7411\n",
      "Epoch 43/200\n",
      "23/23 - 0s - loss: 0.5045 - accuracy: 0.7232\n",
      "Epoch 44/200\n",
      "23/23 - 0s - loss: 0.4901 - accuracy: 0.7589\n",
      "Epoch 45/200\n",
      "23/23 - 0s - loss: 0.4981 - accuracy: 0.7500\n",
      "Epoch 46/200\n",
      "23/23 - 0s - loss: 0.4604 - accuracy: 0.7768\n",
      "Epoch 47/200\n",
      "23/23 - 0s - loss: 0.4531 - accuracy: 0.8839\n",
      "Epoch 48/200\n",
      "23/23 - 0s - loss: 0.4937 - accuracy: 0.7054\n",
      "Epoch 49/200\n",
      "23/23 - 0s - loss: 0.4458 - accuracy: 0.8125\n",
      "Epoch 50/200\n",
      "23/23 - 0s - loss: 0.4809 - accuracy: 0.7232\n",
      "Epoch 51/200\n",
      "23/23 - 0s - loss: 0.4868 - accuracy: 0.7143\n",
      "Epoch 52/200\n",
      "23/23 - 0s - loss: 0.5504 - accuracy: 0.7143\n",
      "Epoch 53/200\n",
      "23/23 - 0s - loss: 0.4702 - accuracy: 0.7857\n",
      "Epoch 54/200\n",
      "23/23 - 0s - loss: 0.4188 - accuracy: 0.8125\n",
      "Epoch 55/200\n",
      "23/23 - 0s - loss: 0.4311 - accuracy: 0.7946\n",
      "Epoch 56/200\n",
      "23/23 - 0s - loss: 0.4282 - accuracy: 0.8304\n",
      "Epoch 57/200\n",
      "23/23 - 0s - loss: 0.4399 - accuracy: 0.7768\n",
      "Epoch 58/200\n",
      "23/23 - 0s - loss: 0.4092 - accuracy: 0.8036\n",
      "Epoch 59/200\n",
      "23/23 - 0s - loss: 0.3910 - accuracy: 0.8393\n",
      "Epoch 60/200\n",
      "23/23 - 0s - loss: 0.4277 - accuracy: 0.7679\n",
      "Epoch 61/200\n",
      "23/23 - 0s - loss: 0.3954 - accuracy: 0.8393\n",
      "Epoch 62/200\n",
      "23/23 - 0s - loss: 0.3918 - accuracy: 0.8482\n",
      "Epoch 63/200\n",
      "23/23 - 0s - loss: 0.4314 - accuracy: 0.7946\n",
      "Epoch 64/200\n",
      "23/23 - 0s - loss: 0.3827 - accuracy: 0.8839\n",
      "Epoch 65/200\n",
      "23/23 - 0s - loss: 0.3803 - accuracy: 0.8661\n",
      "Epoch 66/200\n",
      "23/23 - 0s - loss: 0.3777 - accuracy: 0.8036\n",
      "Epoch 67/200\n",
      "23/23 - 0s - loss: 0.3847 - accuracy: 0.7679\n",
      "Epoch 68/200\n",
      "23/23 - 0s - loss: 0.4626 - accuracy: 0.6964\n",
      "Epoch 69/200\n",
      "23/23 - 0s - loss: 0.4191 - accuracy: 0.7946\n",
      "Epoch 70/200\n",
      "23/23 - 0s - loss: 0.3425 - accuracy: 0.9554\n",
      "Epoch 71/200\n",
      "23/23 - 0s - loss: 0.3537 - accuracy: 0.9107\n",
      "Epoch 72/200\n",
      "23/23 - 0s - loss: 0.3978 - accuracy: 0.7946\n",
      "Epoch 73/200\n",
      "23/23 - 0s - loss: 0.3834 - accuracy: 0.8214\n",
      "Epoch 74/200\n",
      "23/23 - 0s - loss: 0.3632 - accuracy: 0.8750\n",
      "Epoch 75/200\n",
      "23/23 - 0s - loss: 0.3379 - accuracy: 0.9375\n",
      "Epoch 76/200\n",
      "23/23 - 0s - loss: 0.3741 - accuracy: 0.8482\n",
      "Epoch 77/200\n",
      "23/23 - 0s - loss: 0.3995 - accuracy: 0.7679\n",
      "Epoch 78/200\n",
      "23/23 - 0s - loss: 0.3311 - accuracy: 0.9018\n",
      "Epoch 79/200\n",
      "23/23 - 0s - loss: 0.3683 - accuracy: 0.8214\n",
      "Epoch 80/200\n",
      "23/23 - 0s - loss: 0.4422 - accuracy: 0.7500\n",
      "Epoch 81/200\n",
      "23/23 - 0s - loss: 0.3502 - accuracy: 0.8393\n",
      "Epoch 82/200\n",
      "23/23 - 0s - loss: 0.3473 - accuracy: 0.8214\n",
      "Epoch 83/200\n",
      "23/23 - 0s - loss: 0.3416 - accuracy: 0.8393\n",
      "Epoch 84/200\n",
      "23/23 - 0s - loss: 0.3265 - accuracy: 0.9107\n",
      "Epoch 85/200\n",
      "23/23 - 0s - loss: 0.3923 - accuracy: 0.8214\n",
      "Epoch 86/200\n",
      "23/23 - 0s - loss: 0.3707 - accuracy: 0.8214\n",
      "Epoch 87/200\n",
      "23/23 - 0s - loss: 0.3562 - accuracy: 0.8036\n",
      "Epoch 88/200\n",
      "23/23 - 0s - loss: 0.3043 - accuracy: 0.9554\n",
      "Epoch 89/200\n",
      "23/23 - 0s - loss: 0.3685 - accuracy: 0.7589\n",
      "Epoch 90/200\n",
      "23/23 - 0s - loss: 0.3355 - accuracy: 0.9018\n",
      "Epoch 91/200\n",
      "23/23 - 0s - loss: 0.2730 - accuracy: 0.9464\n",
      "Epoch 92/200\n",
      "23/23 - 0s - loss: 0.3243 - accuracy: 0.8661\n",
      "Epoch 93/200\n",
      "23/23 - 0s - loss: 0.3104 - accuracy: 0.8482\n",
      "Epoch 94/200\n",
      "23/23 - 0s - loss: 0.3397 - accuracy: 0.7946\n",
      "Epoch 95/200\n",
      "23/23 - 0s - loss: 0.2800 - accuracy: 0.8839\n",
      "Epoch 96/200\n",
      "23/23 - 0s - loss: 0.2903 - accuracy: 0.9554\n",
      "Epoch 97/200\n",
      "23/23 - 0s - loss: 0.2798 - accuracy: 0.9375\n",
      "Epoch 98/200\n",
      "23/23 - 0s - loss: 0.3186 - accuracy: 0.8393\n",
      "Epoch 99/200\n",
      "23/23 - 0s - loss: 0.2651 - accuracy: 0.9911\n",
      "Epoch 100/200\n",
      "23/23 - 0s - loss: 0.3055 - accuracy: 0.8839\n",
      "Epoch 101/200\n",
      "23/23 - 0s - loss: 0.2803 - accuracy: 0.9107\n",
      "Epoch 102/200\n",
      "23/23 - 0s - loss: 0.3294 - accuracy: 0.8304\n",
      "Epoch 103/200\n",
      "23/23 - 0s - loss: 0.3212 - accuracy: 0.8482\n",
      "Epoch 104/200\n",
      "23/23 - 0s - loss: 0.2660 - accuracy: 0.9286\n",
      "Epoch 105/200\n",
      "23/23 - 0s - loss: 0.2611 - accuracy: 0.9554\n",
      "Epoch 106/200\n",
      "23/23 - 0s - loss: 0.2795 - accuracy: 0.9375\n",
      "Epoch 107/200\n",
      "23/23 - 0s - loss: 0.3451 - accuracy: 0.7857\n",
      "Epoch 108/200\n",
      "23/23 - 0s - loss: 0.2825 - accuracy: 0.8929\n",
      "Epoch 109/200\n",
      "23/23 - 0s - loss: 0.4123 - accuracy: 0.7857\n",
      "Epoch 110/200\n",
      "23/23 - 0s - loss: 0.4385 - accuracy: 0.7857\n",
      "Epoch 111/200\n",
      "23/23 - 0s - loss: 0.3358 - accuracy: 0.8036\n",
      "Epoch 112/200\n",
      "23/23 - 0s - loss: 0.2547 - accuracy: 0.9732\n",
      "Epoch 113/200\n",
      "23/23 - 0s - loss: 0.2365 - accuracy: 0.9732\n",
      "Epoch 114/200\n",
      "23/23 - 0s - loss: 0.2565 - accuracy: 0.9464\n",
      "Epoch 115/200\n",
      "23/23 - 0s - loss: 0.2276 - accuracy: 0.9732\n",
      "Epoch 116/200\n",
      "23/23 - 0s - loss: 0.2723 - accuracy: 0.8839\n",
      "Epoch 117/200\n",
      "23/23 - 0s - loss: 0.2628 - accuracy: 0.9375\n",
      "Epoch 118/200\n",
      "23/23 - 0s - loss: 0.2507 - accuracy: 0.9375\n",
      "Epoch 119/200\n",
      "23/23 - 0s - loss: 0.2601 - accuracy: 0.9196\n",
      "Epoch 120/200\n",
      "23/23 - 0s - loss: 0.2523 - accuracy: 0.8571\n",
      "Epoch 121/200\n",
      "23/23 - 0s - loss: 0.2483 - accuracy: 0.9375\n",
      "Epoch 122/200\n",
      "23/23 - 0s - loss: 0.3628 - accuracy: 0.8036\n",
      "Epoch 123/200\n",
      "23/23 - 0s - loss: 0.2909 - accuracy: 0.8482\n",
      "Epoch 124/200\n",
      "23/23 - 0s - loss: 0.2237 - accuracy: 0.9821\n",
      "Epoch 125/200\n",
      "23/23 - 0s - loss: 0.2277 - accuracy: 0.9554\n",
      "Epoch 126/200\n",
      "23/23 - 0s - loss: 0.2231 - accuracy: 0.9643\n",
      "Epoch 127/200\n",
      "23/23 - 0s - loss: 0.2311 - accuracy: 0.9464\n",
      "Epoch 128/200\n",
      "23/23 - 0s - loss: 0.2601 - accuracy: 0.9107\n",
      "Epoch 129/200\n",
      "23/23 - 0s - loss: 0.1910 - accuracy: 0.9911\n",
      "Epoch 130/200\n",
      "23/23 - 0s - loss: 0.3150 - accuracy: 0.8304\n",
      "Epoch 131/200\n",
      "23/23 - 0s - loss: 0.2456 - accuracy: 0.9286\n",
      "Epoch 132/200\n",
      "23/23 - 0s - loss: 0.2038 - accuracy: 0.9821\n",
      "Epoch 133/200\n",
      "23/23 - 0s - loss: 0.2129 - accuracy: 0.9732\n",
      "Epoch 134/200\n",
      "23/23 - 0s - loss: 0.2074 - accuracy: 0.9643\n",
      "Epoch 135/200\n",
      "23/23 - 0s - loss: 0.2317 - accuracy: 0.9196\n",
      "Epoch 136/200\n",
      "23/23 - 0s - loss: 0.2019 - accuracy: 0.9821\n",
      "Epoch 137/200\n",
      "23/23 - 0s - loss: 0.2164 - accuracy: 0.9821\n",
      "Epoch 138/200\n",
      "23/23 - 0s - loss: 0.2051 - accuracy: 0.9732\n",
      "Epoch 139/200\n",
      "23/23 - 0s - loss: 0.2004 - accuracy: 0.9732\n",
      "Epoch 140/200\n",
      "23/23 - 0s - loss: 0.2615 - accuracy: 0.8839\n",
      "Epoch 141/200\n",
      "23/23 - 0s - loss: 0.1927 - accuracy: 0.9821\n",
      "Epoch 142/200\n",
      "23/23 - 0s - loss: 0.1899 - accuracy: 0.9464\n",
      "Epoch 143/200\n",
      "23/23 - 0s - loss: 0.2483 - accuracy: 0.8929\n",
      "Epoch 144/200\n",
      "23/23 - 0s - loss: 0.1999 - accuracy: 0.9911\n",
      "Epoch 145/200\n",
      "23/23 - 0s - loss: 0.1925 - accuracy: 0.9821\n",
      "Epoch 146/200\n",
      "23/23 - 0s - loss: 0.1940 - accuracy: 0.9286\n",
      "Epoch 147/200\n",
      "23/23 - 0s - loss: 0.2881 - accuracy: 0.8482\n",
      "Epoch 148/200\n",
      "23/23 - 0s - loss: 0.2496 - accuracy: 0.8839\n",
      "Epoch 149/200\n",
      "23/23 - 0s - loss: 0.1889 - accuracy: 0.9464\n",
      "Epoch 150/200\n",
      "23/23 - 0s - loss: 0.2167 - accuracy: 0.9286\n",
      "Epoch 151/200\n",
      "23/23 - 0s - loss: 0.1866 - accuracy: 0.9643\n",
      "Epoch 152/200\n",
      "23/23 - 0s - loss: 0.1942 - accuracy: 0.9643\n",
      "Epoch 153/200\n",
      "23/23 - 0s - loss: 0.1711 - accuracy: 0.9821\n",
      "Epoch 154/200\n",
      "23/23 - 0s - loss: 0.1934 - accuracy: 0.9375\n",
      "Epoch 155/200\n",
      "23/23 - 0s - loss: 0.1919 - accuracy: 0.9464\n",
      "Epoch 156/200\n",
      "23/23 - 0s - loss: 0.2362 - accuracy: 0.8929\n",
      "Epoch 157/200\n",
      "23/23 - 0s - loss: 0.1774 - accuracy: 0.9732\n",
      "Epoch 158/200\n",
      "23/23 - 0s - loss: 0.1834 - accuracy: 0.9643\n",
      "Epoch 159/200\n",
      "23/23 - 0s - loss: 0.1728 - accuracy: 0.9911\n",
      "Epoch 160/200\n",
      "23/23 - 0s - loss: 0.2091 - accuracy: 0.9196\n",
      "Epoch 161/200\n",
      "23/23 - 0s - loss: 0.1670 - accuracy: 0.9732\n",
      "Epoch 162/200\n",
      "23/23 - 0s - loss: 0.1900 - accuracy: 0.9643\n",
      "Epoch 163/200\n",
      "23/23 - 0s - loss: 0.1778 - accuracy: 0.9554\n",
      "Epoch 164/200\n",
      "23/23 - 0s - loss: 0.1929 - accuracy: 0.9643\n",
      "Epoch 165/200\n",
      "23/23 - 0s - loss: 0.2244 - accuracy: 0.9018\n",
      "Epoch 166/200\n",
      "23/23 - 0s - loss: 0.1646 - accuracy: 0.9821\n",
      "Epoch 167/200\n",
      "23/23 - 0s - loss: 0.1811 - accuracy: 0.9554\n",
      "Epoch 168/200\n",
      "23/23 - 0s - loss: 0.1666 - accuracy: 0.9464\n",
      "Epoch 169/200\n",
      "23/23 - 0s - loss: 0.1789 - accuracy: 0.9554\n",
      "Epoch 170/200\n",
      "23/23 - 0s - loss: 0.1766 - accuracy: 0.9732\n",
      "Epoch 171/200\n",
      "23/23 - 0s - loss: 0.1710 - accuracy: 0.9732\n",
      "Epoch 172/200\n",
      "23/23 - 0s - loss: 0.1662 - accuracy: 0.9732\n",
      "Epoch 173/200\n",
      "23/23 - 0s - loss: 0.2090 - accuracy: 0.9107\n",
      "Epoch 174/200\n",
      "23/23 - 0s - loss: 0.1924 - accuracy: 0.9554\n",
      "Epoch 175/200\n",
      "23/23 - 0s - loss: 0.1785 - accuracy: 0.9464\n",
      "Epoch 176/200\n",
      "23/23 - 0s - loss: 0.2294 - accuracy: 0.9107\n",
      "Epoch 177/200\n",
      "23/23 - 0s - loss: 0.1547 - accuracy: 0.9643\n",
      "Epoch 178/200\n",
      "23/23 - 0s - loss: 0.1616 - accuracy: 0.9732\n",
      "Epoch 179/200\n",
      "23/23 - 0s - loss: 0.1598 - accuracy: 0.9732\n",
      "Epoch 180/200\n",
      "23/23 - 0s - loss: 0.1607 - accuracy: 0.9554\n",
      "Epoch 181/200\n",
      "23/23 - 0s - loss: 0.1655 - accuracy: 0.9732\n",
      "Epoch 182/200\n",
      "23/23 - 0s - loss: 0.1629 - accuracy: 0.9821\n",
      "Epoch 183/200\n",
      "23/23 - 0s - loss: 0.1554 - accuracy: 0.9911\n",
      "Epoch 184/200\n",
      "23/23 - 0s - loss: 0.1478 - accuracy: 0.9821\n",
      "Epoch 185/200\n",
      "23/23 - 0s - loss: 0.1537 - accuracy: 0.9732\n",
      "Epoch 186/200\n",
      "23/23 - 0s - loss: 0.1886 - accuracy: 0.9464\n",
      "Epoch 187/200\n",
      "23/23 - 0s - loss: 0.1935 - accuracy: 0.9107\n",
      "Epoch 188/200\n",
      "23/23 - 0s - loss: 0.1394 - accuracy: 0.9911\n",
      "Epoch 189/200\n",
      "23/23 - 0s - loss: 0.1554 - accuracy: 0.9643\n",
      "Epoch 190/200\n",
      "23/23 - 0s - loss: 0.1446 - accuracy: 0.9821\n",
      "Epoch 191/200\n",
      "23/23 - 0s - loss: 0.1829 - accuracy: 0.9375\n",
      "Epoch 192/200\n",
      "23/23 - 0s - loss: 0.1627 - accuracy: 0.9732\n",
      "Epoch 193/200\n",
      "23/23 - 0s - loss: 0.1450 - accuracy: 0.9643\n",
      "Epoch 194/200\n",
      "23/23 - 0s - loss: 0.1630 - accuracy: 0.9732\n",
      "Epoch 195/200\n",
      "23/23 - 0s - loss: 0.1366 - accuracy: 0.9732\n",
      "Epoch 196/200\n",
      "23/23 - 0s - loss: 0.1579 - accuracy: 0.9643\n",
      "Epoch 197/200\n",
      "23/23 - 0s - loss: 0.1484 - accuracy: 0.9732\n",
      "Epoch 198/200\n",
      "23/23 - 0s - loss: 0.1422 - accuracy: 0.9643\n",
      "Epoch 199/200\n",
      "23/23 - 0s - loss: 0.1373 - accuracy: 0.9821\n",
      "Epoch 200/200\n",
      "23/23 - 0s - loss: 0.2363 - accuracy: 0.8750\n",
      "2/2 [==============================] - 0s 500us/step - loss: 0.1459 - accuracy: 0.9211\n",
      "Testing accuracy is 0.9210526347160339\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "import numpy as np\n",
    "# import ELM Multiclass classify\n",
    "import sys\n",
    "from elm import MultiClassify\n",
    "\n",
    "# Load data frame\n",
    "df = load_iris()\n",
    "X = df.data \n",
    "y = df.target.reshape(-1, 1)\n",
    "\n",
    "# One Hot encode the class labels\n",
    "encoder = OneHotEncoder(sparse=False)\n",
    "y = encoder.fit_transform(y)\n",
    "\n",
    "# Split data frame\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25)\n",
    "\n",
    "# Create classify\n",
    "elm = MultiClassify(hidden_layer=['relu'], output_layer='softmax') \n",
    "\n",
    "input_dim = X.shape[1] # number of features\n",
    "elm.buildNetwork(input_dim)\n",
    "\n",
    "elm.fit(X_train, y_train)\n",
    "\n",
    "print (\"Testing accuracy is {}\".format(elm.evaluete(X_test, y_test)[1]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/2 [==============>...............] - ETA: 0s - loss: 0.1634 - accuracy: 0.9062WARNING:tensorflow:Callbacks method `on_test_batch_end` is slow compared to the batch time (batch time: 0.0000s vs `on_test_batch_end` time: 0.0010s). Check your callbacks.\n",
      "2/2 [==============================] - 0s 1ms/step - loss: 0.1459 - accuracy: 0.9211\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.9210526347160339"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "elm.evaluete(X_test, y_test)[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "579282893146967bc5d17141c601e8a8b8ad4a0a2a5a6fe1c87b1000077400d7"
  },
  "kernelspec": {
   "display_name": "Python 3.8.8 ('base')",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}