3 layer FNN

3 layer FNN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Error:0.427202388574\n",
      "Error:0.00806482848555\n",
      "Error:0.00526593550698\n",
      "Error:0.00415712122415\n",
      "Error:0.00352802988241\n",
      "Error:0.00311138599353\n",
      "Error:0.00281016648331\n",
      "Error:0.00257969133595\n",
      "Error:0.00239619398074\n",
      "Error:0.00224573335481\n",
      "[[  5.73112665e-03]\n",
      " [  9.97614085e-01]\n",
      " [  1.55373260e-03]\n",
      " [  1.24248700e-03]\n",
      " [  1.75679361e-03]\n",
      " [  9.97699121e-01]\n",
      " [  6.00341783e-04]\n",
      " [  7.05775684e-04]\n",
      " [  9.97748118e-01]\n",
      " [  9.97748118e-01]\n",
      " [  5.84046510e-04]\n",
      " [  9.97691529e-01]\n",
      " [  9.97739697e-01]\n",
      " [  9.97474914e-01]\n",
      " [  9.97474914e-01]\n",
      " [  9.97474914e-01]\n",
      " [  9.97474914e-01]]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "x = np.array([\n",
    "    [0,0,0,0],\n",
    "    [0,0,0,1],\n",
    "    [0,0,1,0],\n",
    "    [0,1,0,0],\n",
    "    [1,0,0,0],\n",
    "    [0,0,1,1],\n",
    "    [0,1,1,0],\n",
    "    [1,1,0,0],\n",
    "    [1,0,0,1],\n",
    "    [1,0,0,1],\n",
    "    [1,1,1,0],\n",
    "    [1,1,0,1],\n",
    "    [1,0,1,1],\n",
    "    [1,1,1,1],\n",
    "    [1,1,1,1],\n",
    "    [1,1,1,1],\n",
    "    [1,1,1,1]\n",
    "    ])\n",
    "\n",
    "\n",
    "y = np.array([\n",
    "    [0, 1, 0, 0, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1]\n",
    "    ]).T\n",
    "\n",
    "w = np.random.random((4, 4))\n",
    "w2 = np.random.random((4, 4))\n",
    "w3 = np.random.random((4, 1))\n",
    "\n",
    "for j in range(10000):\n",
    "    a2 = 1/(1 + np.exp(-(np.dot(x, w))))\n",
    "    a3 = 1/(1 + np.exp(-(np.dot(a2, w2))))\n",
    "    a4 = 1/(1 + np.exp(-(np.dot(a3, w3))))\n",
    "    a4delta = (y - a4) * (a4 * (1 - a4))\n",
    "    a3delta = a4delta.dot(w3.T) * (a3 * (1 - a3))\n",
    "    a2delta = a3delta.dot(w2.T) * (a2 * (1 - a2))\n",
    "    if (j% 1000) == 0:\n",
    "        print (\"Error:\" + str(np.mean(np.abs(y - a4))))\n",
    "    w3 += a3.T.dot(a4delta)\n",
    "    w2 += a2.T.dot(a3delta)\n",
    "    w += x.T.dot(a2delta)\n",
    "print(a4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Actual value:  0. Predicted value: 0.0057',\n",
       " 'Actual value:  1. Predicted value: 0.9976',\n",
       " 'Actual value:  0. Predicted value: 0.0016',\n",
       " 'Actual value:  0. Predicted value: 0.0012',\n",
       " 'Actual value:  0. Predicted value: 0.0018',\n",
       " 'Actual value:  1. Predicted value: 0.9977',\n",
       " 'Actual value:  0. Predicted value: 0.0006',\n",
       " 'Actual value:  0. Predicted value: 0.0007',\n",
       " 'Actual value:  1. Predicted value: 0.9977',\n",
       " 'Actual value:  1. Predicted value: 0.9977',\n",
       " 'Actual value:  0. Predicted value: 0.0006',\n",
       " 'Actual value:  1. Predicted value: 0.9977',\n",
       " 'Actual value:  1. Predicted value: 0.9977',\n",
       " 'Actual value:  1. Predicted value: 0.9975',\n",
       " 'Actual value:  1. Predicted value: 0.9975',\n",
       " 'Actual value:  1. Predicted value: 0.9975',\n",
       " 'Actual value:  1. Predicted value: 0.9975']"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "['Actual value:  ' + str(i[0][0]) + '. Predicted value: ' + str(np.round(i[1][0], 4)) for i in list(zip(y, a4))]"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [Root]",
   "language": "python",
   "name": "Python [Root]"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}