torch2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x12379e190>]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 随机生成训练集 x 和标签 y\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import random\n",
    "\n",
    "\n",
    "w = 2\n",
    "b = 3\n",
    "xlim = [-10, 10]\n",
    "x_train = np.random.randint(xlim[0], xlim[1], 30)\n",
    "y_train = [w * x + b + random.randint(0, 2) for x in x_train]\n",
    "# bo 表示蓝色圆点, 可以指定线条颜色和形状\n",
    "plt.plot(x_train, y_train, 'bo')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Parameter containing:\n",
      "tensor([1.9658], requires_grad=True) Parameter containing:\n",
      "tensor([3.4545], requires_grad=True)\n",
      "w Parameter containing:\n",
      "tensor([1.9658], requires_grad=True)\n",
      "b Parameter containing:\n",
      "tensor([3.4545], requires_grad=True)\n",
      "Parameter containing:\n",
      "tensor([1.9658], requires_grad=True)\n",
      "Parameter containing:\n",
      "tensor([3.4545], requires_grad=True)\n",
      "Parameter containing:\n",
      "tensor([1.9658], requires_grad=True)\n",
      "Parameter containing:\n",
      "tensor([3.4545], requires_grad=True)\n"
     ]
    }
   ],
   "source": [
    "# 训练回归模型 y = wx + b\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import os\n",
    "\n",
    "\n",
    "# nn.Module 是所有神经网络模块的基类(Base class of all neural network modules)\n",
    "# 我们的模型应该继承这个类(Your models should also subclass this class)\n",
    "class LinearModel(nn.Module):\n",
    "    # __init__ 是类的构造函数, 用于初始化一些参数(Construct function), 必须实现\n",
    "    def __init__(self):\n",
    "        # super(LinearModel, self).__init__()\n",
    "        super().__init__()\n",
    "        self.w = nn.Parameter(torch.randn(1))\n",
    "        self.b = nn.Parameter(torch.randn(1))\n",
    "\n",
    "    # forward 定义了前向传播的运算(如何计算输出), 必须重写. Module类中的__call__方法会调用forward方法\n",
    "    def forward(self, x):\n",
    "        return self.w * x + self.b\n",
    "\n",
    "\n",
    "# 创建模型\n",
    "model = LinearModel()\n",
    "# 定义优化器 随机梯度下降, weight_decay表示L2正则化, momentum表示动量\n",
    "optimizer = optim.SGD(model.parameters(), lr=1e-4,\n",
    "                      weight_decay=1e-2, momentum=0.9)\n",
    "\n",
    "y_train = torch.tensor(y_train, dtype=torch.float32)\n",
    "epoch = 1000\n",
    "for _ in range(epoch):\n",
    "    input = torch.from_numpy(x_train)\n",
    "    # 前向传播\n",
    "    output = model(input)\n",
    "    # 计算损失\n",
    "    loss = nn.MSELoss()(output, y_train)\n",
    "    # 梯度清零\n",
    "    optimizer.zero_grad()\n",
    "    # 反向传播\n",
    "    loss.backward()\n",
    "    # 更新参数\n",
    "    optimizer.step()\n",
    "\n",
    "# 打印训练后的参数\n",
    "print(model.w, model.b)\n",
    "for name, param in model.named_parameters():\n",
    "    print(name, param)\n",
    "\n",
    "if not os.path.exists('./out'):\n",
    "    os.mkdir('./out')\n",
    "\n",
    "# 保存模型参数\n",
    "torch.save(model.state_dict(), './out/linear_model.pth')\n",
    "\n",
    "# 加载模型参数\n",
    "linear_model = LinearModel()\n",
    "linear_model.load_state_dict(torch.load('./out/linear_model.pth'))\n",
    "# 设置为评估模式\n",
    "linear_model.eval()\n",
    "for param in linear_model.parameters():\n",
    "    print(param)\n",
    "\n",
    "# 保存整个模型\n",
    "torch.save(model, './out/linear_model_with_arc.pth')\n",
    "# 加载整个模型, 不需要创建模型了\n",
    "linear_model2 = torch.load('./out/linear_model_with_arc.pth')\n",
    "for param in linear_model2.parameters():\n",
    "    print(param)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}