dl_ex_simple_checkpnt.ipynb

dl_ex_simple_checkpnt.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyMq7ocsn62ZyQTnBQooLG0A",
      "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/gist/dsaint31x/f18e4f3ee942f00da19ba32b7a8beb75/dl_ex_simple_checkpnt.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "A1KUdMU6TakH",
        "outputId": "aada3f24-a416-4316-a5a6-d01759665420"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20, Loss: 0.9961\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 2/20, Loss: 0.9796\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 3/20, Loss: 0.9651\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 4/20, Loss: 0.9526\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 5/20, Loss: 0.9400\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 6/20, Loss: 0.9310\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 7/20, Loss: 0.9175\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 8/20, Loss: 0.9088\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 9/20, Loss: 0.9000\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 10/20, Loss: 0.8908\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 11/20, Loss: 0.8837\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 12/20, Loss: 0.8762\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 13/20, Loss: 0.8696\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 14/20, Loss: 0.8637\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 15/20, Loss: 0.8582\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 16/20, Loss: 0.8517\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 17/20, Loss: 0.8462\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 18/20, Loss: 0.8412\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 19/20, Loss: 0.8368\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Epoch 20/20, Loss: 0.8309\n",
            "Model saved to ./checkpoints/best_model.pth\n",
            "Checkpoint loaded. Training will resume from epoch 20 with loss 0.8309\n"
          ]
        }
      ],
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim\n",
        "from torch.utils.data import DataLoader, TensorDataset\n",
        "import os\n",
        "\n",
        "# 예제 모델 정의\n",
        "class SimpleModel(nn.Module):\n",
        "    def __init__(self):\n",
        "        super(SimpleModel, self).__init__()\n",
        "        self.fc1 = nn.Linear(10, 20)\n",
        "        self.relu = nn.ReLU()\n",
        "        self.fc2 = nn.Linear(20, 1)\n",
        "\n",
        "    def forward(self, x):\n",
        "        x = self.fc1(x)\n",
        "        x = self.relu(x)\n",
        "        x = self.fc2(x)\n",
        "        return x\n",
        "\n",
        "# 데이터셋 생성\n",
        "x_train = torch.randn(100, 10)\n",
        "y_train = torch.randn(100, 1)\n",
        "dataset = TensorDataset(x_train, y_train)\n",
        "dataloader = DataLoader(dataset, batch_size=32, shuffle=True)\n",
        "\n",
        "# 모델, 손실 함수 및 옵티마이저 초기화\n",
        "model = SimpleModel()\n",
        "criterion = nn.MSELoss()\n",
        "optimizer = optim.Adam(model.parameters(), lr=0.001)\n",
        "\n",
        "# 체크포인트 디렉토리 생성\n",
        "checkpoint_dir = './checkpoints'\n",
        "os.makedirs(checkpoint_dir, exist_ok=True)\n",
        "\n",
        "# 최상의 손실 값 초기화\n",
        "best_loss = float('inf')\n",
        "\n",
        "# 훈련 루프\n",
        "num_epochs = 5\n",
        "\n",
        "for epoch in range(num_epochs):\n",
        "    model.train()\n",
        "    running_loss = 0.0\n",
        "    for inputs, targets in dataloader:\n",
        "        optimizer.zero_grad()\n",
        "        outputs = model(inputs)\n",
        "        loss = criterion(outputs, targets)\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "        running_loss += loss.item() * inputs.size(0)\n",
        "\n",
        "    epoch_loss = running_loss / len(dataloader.dataset)\n",
        "    print(f'Epoch {epoch+1}/{num_epochs}, Loss: {epoch_loss:.4f}')\n",
        "\n",
        "    # 손실이 낮아질 때마다 모델 저장\n",
        "    if epoch_loss < best_loss:\n",
        "        best_loss = epoch_loss\n",
        "        checkpoint_path = os.path.join(checkpoint_dir, 'best_model.pth')\n",
        "        torch.save({\n",
        "            'epoch': epoch + 1,\n",
        "            'model_state_dict': model.state_dict(),\n",
        "            'optimizer_state_dict': optimizer.state_dict(),\n",
        "            'loss': best_loss,\n",
        "        }, checkpoint_path)\n",
        "        print(f'Model saved to {checkpoint_path}')\n",
        "\n",
        "# 저장된 모델 로드\n",
        "checkpoint = torch.load(checkpoint_path)\n",
        "model.load_state_dict(checkpoint['model_state_dict'])\n",
        "optimizer.load_state_dict(checkpoint['optimizer_state_dict'])\n",
        "\n",
        "start_epoch = checkpoint['epoch']\n",
        "best_loss = checkpoint['loss']\n",
        "\n",
        "print(f'Checkpoint loaded. Training will resume from epoch {start_epoch} with loss {best_loss:.4f}')\n"
      ]
    }
  ]
}