torch_learning.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyNEjZf+Sazf5br/mMmoHO7R",
      "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/isque03/deeplearning_colab/blob/main/torch_learning.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "0jy7MkJ97JR9"
      },
      "outputs": [],
      "source": [
        "import torch\n",
        "import torch.nn"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "\n",
        "# Define the model\n",
        "class LinearRegression(nn.Module):\n",
        "    def __init__(self):\n",
        "        super().__init__()\n",
        "        self.linear = nn.Linear(1, 1)\n",
        "\n",
        "    def forward(self, x):\n",
        "        return self.linear(x)\n",
        "\n",
        "# Create the model and set the weights\n",
        "model = LinearRegression()\n",
        "w = torch.tensor([[2.0]], requires_grad=True)\n",
        "b = torch.tensor([[1.0]], requires_grad=True)\n",
        "model.linear.weight = nn.Parameter(w)\n",
        "model.linear.bias = nn.Parameter(b)\n",
        "\n",
        "# Define a loss function and an optimizer\n",
        "criterion = nn.MSELoss()\n",
        "optimizer = torch.optim.SGD(model.parameters(), lr=0.01)\n",
        "\n",
        "# Generate some data\n",
        "np.random.seed(42)\n",
        "x_values = np.random.rand(100, 1)\n",
        "y_values = 2 * x_values + 1 + np.random.randn(100, 1)\n",
        "\n",
        "# Convert data to PyTorch tensors\n",
        "x_train = torch.from_numpy(x_values).float()\n",
        "y_train = torch.from_numpy(y_values).float()\n",
        "\n",
        "# Train the model\n",
        "num_epochs = 100\n",
        "for epoch in range(num_epochs):\n",
        "    # Forward pass\n",
        "    y_pred = model(x_train)\n",
        "    loss = criterion(y_pred, y_train)\n",
        "\n",
        "    # Backward and optimize\n",
        "    optimizer.zero_grad()\n",
        "    loss.backward()\n",
        "    optimizer.step()\n",
        "\n",
        "# Print the final weights\n",
        "print(f'w = {model.linear.weight.item():.4f}, b = {model.linear.bias.item():.4f}')\n",
        "\n",
        "# Plot the results\n",
        "sns.scatterplot(x_values.flatten(), y_values.flatten())\n",
        "plt.plot(x_values, y_pred.data.numpy(), 'r-')\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 341
        },
        "id": "IQhzXAv37Ruq",
        "outputId": "e21d2238-e141-46e2-8207-f0b57da4e705"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "w = 1.9337, b = 1.0174\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.8/dist-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}