Created using Colaboratory

Author: alekyaarra

Neural_Networks_using_Pytorch.ipynb

@@ -0,0 +1,197 @@
+{
+  "nbformat": 4,
+  "nbformat_minor": 0,
+  "metadata": {
+    "colab": {
+      "provenance": [],
+      "authorship_tag": "ABX9TyOqT4fV0HqqEjJ2PIVhiM/p",
+      "include_colab_link": true
+    },
+    "kernelspec": {
+      "name": "python3",
+      "display_name": "Python 3"
+    },
+    "language_info": {
+      "name": "python"
+    },
+    "accelerator": "GPU",
+    "gpuClass": "standard"
+  },
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "view-in-github",
+        "colab_type": "text"
+      },
+      "source": [
+        "<a href=\"https://colab.research.google.com/github/smlra-kjsce/Pytorch-101-2023/blob/main/Neural_Networks_using_Pytorch.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "import torch"
+      ],
+      "metadata": {
+        "id": "tZeF_iz7llZh"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "RqcIgy3Iy5GF",
+        "outputId": "67bc0af0-9d28-4f5d-ff99-9357f67ec58f"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "no of features:  4\n",
+            "tensor([[1., 0., 1., 0.],\n",
+            "        [1., 0., 1., 1.],\n",
+            "        [0., 1., 0., 1.]]) \n",
+            "\n",
+            "tensor([[1.],\n",
+            "        [1.],\n",
+            "        [0.]])\n"
+          ]
+        }
+      ],
+      "source": [
+        "\n",
+        "#Input tensor\n",
+        "X = torch.Tensor([[1,0,1,0],[1,0,1,1],[0,1,0,1]])\n",
+        "\n",
+        "print(\"no of features: \", X.shape[1])\n",
+        "#Output\n",
+        "y = torch.Tensor([[1],[1],[0]])\n",
+        "\n",
+        "\n",
+        "print(X, '\\n')\n",
+        "print(y)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "#Sigmoid Function\n",
+        "def sigmoid (x):\n",
+        "    return 1/(1 + torch.exp(-x))\n",
+        "\n",
+        "#Derivative of Sigmoid Function/\n",
+        "def derivatives_sigmoid(x):\n",
+        "    return sigmoid(x) * (1 - sigmoid(x))"
+      ],
+      "metadata": {
+        "id": "nLfTl3Yny53Q"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "#Variable initialization\n",
+        "\n",
+        "epoch=7000 #Setting training iterations\n",
+        "lr=0.1 #Setting learning rate\n",
+        "\n",
+        "inputlayer_neurons = X.shape[1] #number of features in data set\n",
+        "hiddenlayer_neurons = 3 #number of hidden layer neurons\n",
+        "output_neurons = 1 #number of neurons in output layer\n",
+        "\n",
+        "#weight and bias initialization\n",
+        "wh=torch.randn(inputlayer_neurons, hiddenlayer_neurons).type(torch.FloatTensor)\n",
+        "bh=torch.randn(1, hiddenlayer_neurons).type(torch.FloatTensor)\n",
+        "\n",
+        "wout=torch.randn(hiddenlayer_neurons, output_neurons)\n",
+        "bout=torch.randn(1, output_neurons)"
+      ],
+      "metadata": {
+        "id": "0rqFB4BHzDV2"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "for i in range(epoch):\n",
+        "    #Forward Propogation\n",
+        "    hidden_layer_input1 = torch.mm(X, wh)\n",
+        "    hidden_layer_input = hidden_layer_input1 + bh\n",
+        "    hidden_layer_activations = sigmoid(hidden_layer_input)\n",
+        "\n",
+        "    output_layer_input1 = torch.mm(hidden_layer_activations, wout)\n",
+        "    output_layer_input = output_layer_input1 + bout\n",
+        "    output = sigmoid(output_layer_input)\n",
+        "\n",
+        "    #Backpropagation\n",
+        "    E = y-output\n",
+        "    slope_output_layer = derivatives_sigmoid(output)\n",
+        "    slope_hidden_layer = derivatives_sigmoid(hidden_layer_activations)\n",
+        "    d_output = E * slope_output_layer\n",
+        "    Error_at_hidden_layer = torch.mm(d_output, wout.t())\n",
+        "    d_hiddenlayer = Error_at_hidden_layer * slope_hidden_layer\n",
+        "    wout += torch.mm(hidden_layer_activations.t(), d_output) *lr\n",
+        "    bout += d_output.sum() *lr\n",
+        "    wh += torch.mm(X.t(), d_hiddenlayer) *lr\n",
+        "    bh += d_output.sum() *lr"
+      ],
+      "metadata": {
+        "id": "HqxBtOtCzFbf"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "print('actual :\\n', y, '\\n')\n",
+        "print('predicted :\\n', output)"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "Llt_Alt0zHq6",
+        "outputId": "41ab11c6-4a95-416d-b414-85d379a2b724"
+      },
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "actual :\n",
+            " tensor([[1.],\n",
+            "        [1.],\n",
+            "        [0.]]) \n",
+            "\n",
+            "predicted :\n",
+            " tensor([[0.9965],\n",
+            "        [0.9964],\n",
+            "        [0.0056]])\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [],
+      "metadata": {
+        "id": "M4ti0eLazJRb"
+      },
+      "execution_count": null,
+      "outputs": []
+    }
+  ]
+}