01_basics.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "01_basics.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/subhobrata/PyTorchZeroAll/blob/master/01_basics.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "metadata": {
        "id": "-kZ-VCBfRuYB",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "AuM9dOqMVZI5",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "x_data = [1.0,2.0,3.0]\n",
        "y_data = [2.0,4.0,6.0]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "af3hp4TAVlto",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def forward(x):\n",
        "    return x * w"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "AYW0Mm83Vx4p",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Loss function\n",
        "def loss(x,y):\n",
        "  y_pred = forward(x)\n",
        "  return (y_pred - y) * (y_pred - y)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "hqtR88IvWLPo",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "w_list = []\n",
        "mse_list = []"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "7fBgbAdoWR7g",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 3846
        },
        "outputId": "aaed2c8f-c0db-4a27-a7a6-7c0b66af9e50"
      },
      "cell_type": "code",
      "source": [
        "for w in np.arange(0.0, 4.1, 0.1):\n",
        "    print(\"w=\", w)\n",
        "    l_sum = 0\n",
        "    for x_val, y_val in zip(x_data, y_data):\n",
        "        y_pred_val = forward(x_val)\n",
        "        l = loss(x_val, y_val)\n",
        "        l_sum += l\n",
        "        print(\"\\t\", x_val, y_val, y_pred_val, l)\n",
        "    print(\"MSE=\", l_sum / 3)\n",
        "    w_list.append(w)\n",
        "    mse_list.append(l_sum / 3)\n",
        "\n",
        "plt.plot(w_list, mse_list)\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('w')\n",
        "plt.show()\n",
        "    "
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "w= 0.0\n",
            "\t 1.0 2.0 0.0 4.0\n",
            "\t 2.0 4.0 0.0 16.0\n",
            "\t 3.0 6.0 0.0 36.0\n",
            "MSE= 18.666666666666668\n",
            "w= 0.1\n",
            "\t 1.0 2.0 0.1 3.61\n",
            "\t 2.0 4.0 0.2 14.44\n",
            "\t 3.0 6.0 0.30000000000000004 32.49\n",
            "MSE= 16.846666666666668\n",
            "w= 0.2\n",
            "\t 1.0 2.0 0.2 3.24\n",
            "\t 2.0 4.0 0.4 12.96\n",
            "\t 3.0 6.0 0.6000000000000001 29.160000000000004\n",
            "MSE= 15.120000000000003\n",
            "w= 0.30000000000000004\n",
            "\t 1.0 2.0 0.30000000000000004 2.8899999999999997\n",
            "\t 2.0 4.0 0.6000000000000001 11.559999999999999\n",
            "\t 3.0 6.0 0.9000000000000001 26.009999999999998\n",
            "MSE= 13.486666666666665\n",
            "w= 0.4\n",
            "\t 1.0 2.0 0.4 2.5600000000000005\n",
            "\t 2.0 4.0 0.8 10.240000000000002\n",
            "\t 3.0 6.0 1.2000000000000002 23.04\n",
            "MSE= 11.946666666666667\n",
            "w= 0.5\n",
            "\t 1.0 2.0 0.5 2.25\n",
            "\t 2.0 4.0 1.0 9.0\n",
            "\t 3.0 6.0 1.5 20.25\n",
            "MSE= 10.5\n",
            "w= 0.6000000000000001\n",
            "\t 1.0 2.0 0.6000000000000001 1.9599999999999997\n",
            "\t 2.0 4.0 1.2000000000000002 7.839999999999999\n",
            "\t 3.0 6.0 1.8000000000000003 17.639999999999993\n",
            "MSE= 9.146666666666663\n",
            "w= 0.7000000000000001\n",
            "\t 1.0 2.0 0.7000000000000001 1.6899999999999995\n",
            "\t 2.0 4.0 1.4000000000000001 6.759999999999998\n",
            "\t 3.0 6.0 2.1 15.209999999999999\n",
            "MSE= 7.886666666666666\n",
            "w= 0.8\n",
            "\t 1.0 2.0 0.8 1.44\n",
            "\t 2.0 4.0 1.6 5.76\n",
            "\t 3.0 6.0 2.4000000000000004 12.959999999999997\n",
            "MSE= 6.719999999999999\n",
            "w= 0.9\n",
            "\t 1.0 2.0 0.9 1.2100000000000002\n",
            "\t 2.0 4.0 1.8 4.840000000000001\n",
            "\t 3.0 6.0 2.7 10.889999999999999\n",
            "MSE= 5.646666666666666\n",
            "w= 1.0\n",
            "\t 1.0 2.0 1.0 1.0\n",
            "\t 2.0 4.0 2.0 4.0\n",
            "\t 3.0 6.0 3.0 9.0\n",
            "MSE= 4.666666666666667\n",
            "w= 1.1\n",
            "\t 1.0 2.0 1.1 0.8099999999999998\n",
            "\t 2.0 4.0 2.2 3.2399999999999993\n",
            "\t 3.0 6.0 3.3000000000000003 7.289999999999998\n",
            "MSE= 3.779999999999999\n",
            "w= 1.2000000000000002\n",
            "\t 1.0 2.0 1.2000000000000002 0.6399999999999997\n",
            "\t 2.0 4.0 2.4000000000000004 2.5599999999999987\n",
            "\t 3.0 6.0 3.6000000000000005 5.759999999999997\n",
            "MSE= 2.986666666666665\n",
            "w= 1.3\n",
            "\t 1.0 2.0 1.3 0.48999999999999994\n",
            "\t 2.0 4.0 2.6 1.9599999999999997\n",
            "\t 3.0 6.0 3.9000000000000004 4.409999999999998\n",
            "MSE= 2.2866666666666657\n",
            "w= 1.4000000000000001\n",
            "\t 1.0 2.0 1.4000000000000001 0.3599999999999998\n",
            "\t 2.0 4.0 2.8000000000000003 1.4399999999999993\n",
            "\t 3.0 6.0 4.2 3.2399999999999993\n",
            "MSE= 1.6799999999999995\n",
            "w= 1.5\n",
            "\t 1.0 2.0 1.5 0.25\n",
            "\t 2.0 4.0 3.0 1.0\n",
            "\t 3.0 6.0 4.5 2.25\n",
            "MSE= 1.1666666666666667\n",
            "w= 1.6\n",
            "\t 1.0 2.0 1.6 0.15999999999999992\n",
            "\t 2.0 4.0 3.2 0.6399999999999997\n",
            "\t 3.0 6.0 4.800000000000001 1.4399999999999984\n",
            "MSE= 0.746666666666666\n",
            "w= 1.7000000000000002\n",
            "\t 1.0 2.0 1.7000000000000002 0.0899999999999999\n",
            "\t 2.0 4.0 3.4000000000000004 0.3599999999999996\n",
            "\t 3.0 6.0 5.1000000000000005 0.809999999999999\n",
            "MSE= 0.4199999999999995\n",
            "w= 1.8\n",
            "\t 1.0 2.0 1.8 0.03999999999999998\n",
            "\t 2.0 4.0 3.6 0.15999999999999992\n",
            "\t 3.0 6.0 5.4 0.3599999999999996\n",
            "MSE= 0.1866666666666665\n",
            "w= 1.9000000000000001\n",
            "\t 1.0 2.0 1.9000000000000001 0.009999999999999974\n",
            "\t 2.0 4.0 3.8000000000000003 0.0399999999999999\n",
            "\t 3.0 6.0 5.7 0.0899999999999999\n",
            "MSE= 0.046666666666666586\n",
            "w= 2.0\n",
            "\t 1.0 2.0 2.0 0.0\n",
            "\t 2.0 4.0 4.0 0.0\n",
            "\t 3.0 6.0 6.0 0.0\n",
            "MSE= 0.0\n",
            "w= 2.1\n",
            "\t 1.0 2.0 2.1 0.010000000000000018\n",
            "\t 2.0 4.0 4.2 0.04000000000000007\n",
            "\t 3.0 6.0 6.300000000000001 0.09000000000000043\n",
            "MSE= 0.046666666666666835\n",
            "w= 2.2\n",
            "\t 1.0 2.0 2.2 0.04000000000000007\n",
            "\t 2.0 4.0 4.4 0.16000000000000028\n",
            "\t 3.0 6.0 6.6000000000000005 0.36000000000000065\n",
            "MSE= 0.18666666666666698\n",
            "w= 2.3000000000000003\n",
            "\t 1.0 2.0 2.3000000000000003 0.09000000000000016\n",
            "\t 2.0 4.0 4.6000000000000005 0.36000000000000065\n",
            "\t 3.0 6.0 6.9 0.8100000000000006\n",
            "MSE= 0.42000000000000054\n",
            "w= 2.4000000000000004\n",
            "\t 1.0 2.0 2.4000000000000004 0.16000000000000028\n",
            "\t 2.0 4.0 4.800000000000001 0.6400000000000011\n",
            "\t 3.0 6.0 7.200000000000001 1.4400000000000026\n",
            "MSE= 0.7466666666666679\n",
            "w= 2.5\n",
            "\t 1.0 2.0 2.5 0.25\n",
            "\t 2.0 4.0 5.0 1.0\n",
            "\t 3.0 6.0 7.5 2.25\n",
            "MSE= 1.1666666666666667\n",
            "w= 2.6\n",
            "\t 1.0 2.0 2.6 0.3600000000000001\n",
            "\t 2.0 4.0 5.2 1.4400000000000004\n",
            "\t 3.0 6.0 7.800000000000001 3.2400000000000024\n",
            "MSE= 1.6800000000000008\n",
            "w= 2.7\n",
            "\t 1.0 2.0 2.7 0.49000000000000027\n",
            "\t 2.0 4.0 5.4 1.960000000000001\n",
            "\t 3.0 6.0 8.100000000000001 4.410000000000006\n",
            "MSE= 2.2866666666666693\n",
            "w= 2.8000000000000003\n",
            "\t 1.0 2.0 2.8000000000000003 0.6400000000000005\n",
            "\t 2.0 4.0 5.6000000000000005 2.560000000000002\n",
            "\t 3.0 6.0 8.4 5.760000000000002\n",
            "MSE= 2.986666666666668\n",
            "w= 2.9000000000000004\n",
            "\t 1.0 2.0 2.9000000000000004 0.8100000000000006\n",
            "\t 2.0 4.0 5.800000000000001 3.2400000000000024\n",
            "\t 3.0 6.0 8.700000000000001 7.290000000000005\n",
            "MSE= 3.780000000000003\n",
            "w= 3.0\n",
            "\t 1.0 2.0 3.0 1.0\n",
            "\t 2.0 4.0 6.0 4.0\n",
            "\t 3.0 6.0 9.0 9.0\n",
            "MSE= 4.666666666666667\n",
            "w= 3.1\n",
            "\t 1.0 2.0 3.1 1.2100000000000002\n",
            "\t 2.0 4.0 6.2 4.840000000000001\n",
            "\t 3.0 6.0 9.3 10.890000000000004\n",
            "MSE= 5.646666666666668\n",
            "w= 3.2\n",
            "\t 1.0 2.0 3.2 1.4400000000000004\n",
            "\t 2.0 4.0 6.4 5.760000000000002\n",
            "\t 3.0 6.0 9.600000000000001 12.96000000000001\n",
            "MSE= 6.720000000000003\n",
            "w= 3.3000000000000003\n",
            "\t 1.0 2.0 3.3000000000000003 1.6900000000000006\n",
            "\t 2.0 4.0 6.6000000000000005 6.7600000000000025\n",
            "\t 3.0 6.0 9.9 15.210000000000003\n",
            "MSE= 7.886666666666668\n",
            "w= 3.4000000000000004\n",
            "\t 1.0 2.0 3.4000000000000004 1.960000000000001\n",
            "\t 2.0 4.0 6.800000000000001 7.840000000000004\n",
            "\t 3.0 6.0 10.200000000000001 17.640000000000008\n",
            "MSE= 9.14666666666667\n",
            "w= 3.5\n",
            "\t 1.0 2.0 3.5 2.25\n",
            "\t 2.0 4.0 7.0 9.0\n",
            "\t 3.0 6.0 10.5 20.25\n",
            "MSE= 10.5\n",
            "w= 3.6\n",
            "\t 1.0 2.0 3.6 2.5600000000000005\n",
            "\t 2.0 4.0 7.2 10.240000000000002\n",
            "\t 3.0 6.0 10.8 23.040000000000006\n",
            "MSE= 11.94666666666667\n",
            "w= 3.7\n",
            "\t 1.0 2.0 3.7 2.8900000000000006\n",
            "\t 2.0 4.0 7.4 11.560000000000002\n",
            "\t 3.0 6.0 11.100000000000001 26.010000000000016\n",
            "MSE= 13.486666666666673\n",
            "w= 3.8000000000000003\n",
            "\t 1.0 2.0 3.8000000000000003 3.240000000000001\n",
            "\t 2.0 4.0 7.6000000000000005 12.960000000000004\n",
            "\t 3.0 6.0 11.4 29.160000000000004\n",
            "MSE= 15.120000000000005\n",
            "w= 3.9000000000000004\n",
            "\t 1.0 2.0 3.9000000000000004 3.610000000000001\n",
            "\t 2.0 4.0 7.800000000000001 14.440000000000005\n",
            "\t 3.0 6.0 11.700000000000001 32.49000000000001\n",
            "MSE= 16.84666666666667\n",
            "w= 4.0\n",
            "\t 1.0 2.0 4.0 4.0\n",
            "\t 2.0 4.0 8.0 16.0\n",
            "\t 3.0 6.0 12.0 36.0\n",
            "MSE= 18.666666666666668\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFYCAYAAABKymUhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 576x396 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}