#7_StreetlightProblemDeepNeuralNetwork.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as  np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# StreetLight Problem\n",
    "The problem is that in an unknown country the street lights have different signals that is to be decoded so that a stranger can walk or stop according to a specific state of the lights.<br>\n",
    "Here there are three light with their own individual states that collectively can mean to `stop` or to `walk`.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "streetlights = np.array([[1,0,1],\n",
    "                         [0,1,1],\n",
    "                         [0,0,1],\n",
    "                         [1,1,1],\n",
    "                         [0,1,1],\n",
    "                         [1,0,1],])\n",
    "walk_vs_stop = np.array([[0],\n",
    "                         [1],\n",
    "                         [0],\n",
    "                         [1],\n",
    "                         [1],\n",
    "                         [0]])\n",
    "np.random.seed(1)\n",
    "\n",
    "weights1 = np.random.rand(3,3)\n",
    "weights2 = np.random.rand(3,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of streetlights dataset (6, 3)\n",
      "Shape of walk_vs_stop dataset (6, 1)\n",
      "Shape of weights1 (3, 3)\n",
      "Shape of weights2 (3, 1)\n"
     ]
    }
   ],
   "source": [
    "print('Shape of streetlights dataset {shape}'.format(shape=streetlights.shape))\n",
    "print('Shape of walk_vs_stop dataset {shape}'.format(shape=walk_vs_stop.shape))\n",
    "print('Shape of weights1 {shape}'.format(shape=weights1.shape))\n",
    "print('Shape of weights2 {shape}'.format(shape=weights2.shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def relu(x):\n",
    "    return (x > 0) * x\n",
    "def relu2deriv(x):\n",
    "    return (x > 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Intermediate Layer\n",
    "The `intermediate layer` can be thought of a `separate dataset layer` that has been made. The input to the intermediate layer is a culmination of the `real dataset` and `weights`. This creates it's own kind of dataset, that might have a `better correlation` to the `output` that we are looking for."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr = 0.1\n",
    "errors = []\n",
    "weights_graph = []\n",
    "for iteration in range(100):\n",
    "    for index in range(len(streetlights)):\n",
    "        \n",
    "        pred1 = relu(streetlights[index] @ weights1)\n",
    "        pred2 = pred1 @ weights2\n",
    "\n",
    "        error_out = pred2 - walk_vs_stop[index]\n",
    "        error_inter = error_out * weights2\n",
    "\n",
    "        delta_weight_2 = (pred1.T * error_out).reshape(-1,1)\n",
    "        delta_weight_1 = streetlights[index] * error_inter * relu2deriv(pred1)\n",
    "        \n",
    "        weights1 = weights1 - (delta_weight_1)\n",
    "        weights2 = weights2 - (delta_weight_2)\n",
    "    if iteration % 10 == 0:\n",
    "        errors.append(error_out)\n",
    "        weights_graph.append(weights2.flatten())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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"
    }
   ],
   "source": [
    "plt.plot(errors)\n",
    "plt.xlabel('10th epoch')\n",
    "plt.ylabel('Error of Outer Layer')\n",
    "plt.title('Error Graph')\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.8.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}