ANN_Regression.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ANN Regression.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPj920bvqB8KQjU5ExBnNjr",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/sancoap/pytorch1/blob/main/ANN_Regression.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S74P-Ym0mbAl"
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "from mpl_toolkits.mplot3d import Axes3D"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "D4NnaKnOmvGn"
      },
      "source": [
        "#Make a synthetic dataset\n",
        "#random function returns the data uniformly distributed between 0 and 1\n",
        "#we multiple this by 6 and remove 3 to get distribution between -3 and +3?\n",
        "N = 1000\n",
        "X = np.random.random((N,2)) * 6 - 3 #uniformly distributed between -3 and +3\n",
        "Y = np.cos(2*X[:,0]) + np.cos(3*X[:,0])"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "m4kmsHWFnMgw",
        "outputId": "ef9a28eb-6e5b-4801-ac55-67eceeae9cf9"
      },
      "source": [
        "#you cannot play with teh plot in notebook\n",
        "fig = plt.figure()\n",
        "ax = fig.add_subplot(111, projection='3d')\n",
        "ax.scatter(X[:,0], X[:,1], Y)\n"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<mpl_toolkits.mplot3d.art3d.Path3DCollection at 0x7fc61040a690>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xw0E8gnooySW"
      },
      "source": [
        "#in our model we have 2 inputs and 128 nodes in 1 layer and then 1 output function\n",
        "model = nn.Sequential (\n",
        "    nn.Linear(2,128),\n",
        "    nn.ReLU(),\n",
        "    nn.Linear(128,1)\n",
        ")"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FWCtxkrypBGM"
      },
      "source": [
        "criterion = nn.MSELoss()\n",
        "optimizer = torch.optim.Adam(model.parameters(), lr=0.01)"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 358
        },
        "id": "WV_wbWyZqIhe",
        "outputId": "7cb8be78-9a83-4ce6-d937-f86574d38d50"
      },
      "source": [
        "def full_gd(model, criterion, optimizer, X_train, y_train, epochs=1000):\n",
        "  train_losses = np.zeros(epochs)\n",
        "  for it in range(epochs):\n",
        "    optimizer.zero_grad()\n",
        "    #Forward Pass\n",
        "    outputs = model(X_train)\n",
        "    loss = criterion(outputs,y_train)\n",
        "\n",
        "    #backward pass\n",
        "    loss.backward()\n",
        "    optimizer.step()\n",
        "\n",
        "    train_losses[it] = loss.item()\n",
        "    if(it +1) % 50 == 0:\n",
        "      print(f'Epochs {it+1}/{epochs}, Train Loss: {loss.item():.4f} ')\n",
        "  return train_losses\n",
        "\n",
        "X_train = torch.from_numpy(X.astype(np.float32))\n",
        "y_train = torch.from_numpy(Y.astype(np.float32).reshape(-1,1))\n",
        "train_losses = full_gd(model, criterion, optimizer, X_train, y_train)\n",
        "\n"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "error",
          "ename": "RuntimeError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mRuntimeError\u001b[0m                              Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-15-187ac0ff045d>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     18\u001b[0m \u001b[0mX_train\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfrom_numpy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfloat32\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     19\u001b[0m \u001b[0my_train\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfrom_numpy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mY\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfloat32\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 20\u001b[0;31m \u001b[0mtrain_losses\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfull_gd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcriterion\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     21\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m<ipython-input-15-187ac0ff045d>\u001b[0m in \u001b[0;36mfull_gd\u001b[0;34m(model, criterion, optimizer, X_train, y_train, epochs)\u001b[0m\n\u001b[1;32m      4\u001b[0m     \u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzero_grad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m     \u001b[0;31m#Forward Pass\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m     \u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      7\u001b[0m     \u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcriterion\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0my_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    887\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    888\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 889\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    890\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    891\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/container.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    117\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    118\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 119\u001b[0;31m             \u001b[0minput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodule\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    120\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    121\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    887\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    888\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 889\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    890\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    891\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/linear.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m     92\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     93\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mTensor\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0mTensor\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 94\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mF\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlinear\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mweight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbias\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     95\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     96\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mextra_repr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py\u001b[0m in \u001b[0;36mlinear\u001b[0;34m(input, weight, bias)\u001b[0m\n\u001b[1;32m   1751\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mhas_torch_function_variadic\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweight\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1752\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mhandle_torch_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlinear\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweight\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbias\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mbias\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1753\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_C\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_nn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlinear\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbias\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1754\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1755\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mRuntimeError\u001b[0m: mat1 and mat2 shapes cannot be multiplied (1000x2 and 1x128)"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SKqCpWo-r7Is"
      },
      "source": [
        "plt.plot(train_losses)"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}