{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "UKqu0Ep9HEkF"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "X = np.array(([2, 9], [1, 5], [3, 6]), dtype=float) # two inputs [sleep,study]\n",
        "y = np.array(([92], [86], [89]), dtype=float) # one output [Expected % in Exams]\n",
        "X = X/np.amax(X,axis=0) # maximum of X array longitudinally\n",
        "y = y/100"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Sigmoid Function\n",
        "def sigmoid (x):\n",
        "    return 1/(1 + np.exp(-x))\n",
        "\n",
        "\n",
        "#Derivative of Sigmoid Function\n",
        "def derivatives_sigmoid(x):\n",
        "    return x * (1 - x)"
      ],
      "metadata": {
        "id": "7048eh0mHHla"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Variable initialization\n",
        "epoch=5000  #Setting training iterations\n",
        "lr=0.1    #Setting learning rate\n",
        "inputlayer_neurons = 2    #number of features in data set\n",
        "hiddenlayer_neurons = 3   #number of hidden layers neurons\n",
        "output_neurons = 1    #number of neurons at output layer"
      ],
      "metadata": {
        "id": "LlXGsPfUHL3I"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#weight and bias initialization\n",
        "wh=np.random.uniform(size=(inputlayer_neurons,hiddenlayer_neurons)) #weight of the link from input node to hidden node\n",
        "bh=np.random.uniform(size=(1,hiddenlayer_neurons)) # bias of the link from input node to hidden node\n",
        "wout=np.random.uniform(size=(hiddenlayer_neurons,output_neurons)) #weight of the link from hidden node to output node\n",
        "bout=np.random.uniform(size=(1,output_neurons)) #bias of the link from hidden node to output node\n"
      ],
      "metadata": {
        "id": "tbfwqIHpHOUr"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#draws a random range of numbers uniformly of dim x*y\n",
        "for i in range(epoch):\n",
        "  #Forward Propogation\n",
        "    hinp1=np.dot(X,wh)\n",
        "    hinp=hinp1 + bh\n",
        "    hlayer_act = sigmoid(hinp)\n",
        "    outinp1=np.dot(hlayer_act,wout)\n",
        "    outinp= outinp1+ bout\n",
        "    output = sigmoid(outinp)\n",
        "\n",
        "\n",
        "#Backpropagation\n",
        "    EO = y-output\n",
        "    outgrad = derivatives_sigmoid(output)\n",
        "    d_output = EO* outgrad\n",
        "    EH = d_output.dot(wout.T)\n",
        "\n",
        "\n",
        "#how much hidden layer weights contributed to error\n",
        "    hiddengrad = derivatives_sigmoid(hlayer_act)\n",
        "    d_hiddenlayer = EH * hiddengrad\n",
        "\n",
        "\n",
        "# dotproduct of nextlayererror and currentlayerop\n",
        "wout += hlayer_act.T.dot(d_output) *lr\n",
        "  \n",
        "\n",
        "wh += X.T.dot(d_hiddenlayer) *lr\n",
        "print(\"Input: \\n\" + str(X)) \n",
        "print(\"Actual Output: \\n\" + str(y))\n",
        "print(\"Predicted Output: \\n\" ,output)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ArjgUk33HSJn",
        "outputId": "9d4654cc-d8fc-4b51-e5c3-078c0ab557fb"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Input: \n",
            "[[0.66666667 1.        ]\n",
            " [0.33333333 0.55555556]\n",
            " [1.         0.66666667]]\n",
            "Actual Output: \n",
            "[[0.92]\n",
            " [0.86]\n",
            " [0.89]]\n",
            "Predicted Output: \n",
            " [[0.76983028]\n",
            " [0.75596362]\n",
            " [0.77020531]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "1z1LVDOEHZ-g"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}