神经网络简单模型训练.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6e6270f0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ship bird frog ship\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torchvision\n",
    "import torchvision.transforms as transforms\n",
    "import numpy as np\n",
    "import os\n",
    "os.environ[\"KMP_DUPLICATE_LIB_OK\"]=\"TRUE\"\n",
    "\n",
    "transform=transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5,0.5,0.5),(0.5,0.5,0.5))])\n",
    "\n",
    "trainset=torchvision.datasets.CIFAR10(root='./data',train=True,download=True,transform=transform)\n",
    "\n",
    "trainloader=torch.utils.data.DataLoader(trainset,batch_size=4,shuffle=True,num_workers=2)\n",
    "\n",
    "testset=torchvision.datasets.CIFAR10(root='./data',train=False,download=True,transform=transform)\n",
    "\n",
    "testloader=torch.utils.data.DataLoader(testset,batch_size=4,shuffle=False,num_workers=2)\n",
    "\n",
    "classes=('plane','car','bird','cat','deer','dog','frog','horse','ship','truck')\n",
    "\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "#构建展示图片的函数\n",
    "def imshow(img):\n",
    "    img=img/2+0.5\n",
    "    npimg=img.numpy()\n",
    "    plt.imshow(np.transpose(npimg,(1,2,0)))\n",
    "    plt.show()\n",
    "\n",
    "#从数据迭代器中读取一张图片\n",
    "dataiter=iter(trainloader)\n",
    "images,labels=dataiter.next()\n",
    "\n",
    "#展示图片\n",
    "imshow(torchvision.utils.make_grid(images))\n",
    "#打印标签\n",
    "print(''.join('%5s'% classes[labels[j]] for j in range(4)))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "36822e99",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Net(\n",
      "  (conv1): Conv2d(3, 6, kernel_size=(5, 5), stride=(1, 1))\n",
      "  (conv2): Conv2d(6, 16, kernel_size=(5, 5), stride=(1, 1))\n",
      "  (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
      "  (fc1): Linear(in_features=400, out_features=120, bias=True)\n",
      "  (fc2): Linear(in_features=120, out_features=84, bias=True)\n",
      "  (fc3): Linear(in_features=84, out_features=10, bias=True)\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "\n",
    "class Net(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(Net,self).__init__()\n",
    "        #定义两个卷积层\n",
    "        self.conv1=nn.Conv2d(3,6,5)#有RGB的图片第一个参数channel应该为3,这里为了看size的变化才写1的\n",
    "        self.conv2=nn.Conv2d(6,16,5)\n",
    "        #定义池化层\n",
    "        self.pool=nn.MaxPool2d(2,2)\n",
    "        #定义三个全连接层\n",
    "        self.fc1=nn.Linear(16*5*5,120)\n",
    "        self.fc2=nn.Linear(120,84)\n",
    "        self.fc3=nn.Linear(84,10)\n",
    "\n",
    "    def forward(self,x):\n",
    "        x=self.pool(F.relu(self.conv1(x)))\n",
    "        x=self.pool(F.relu(self.conv2(x)))\n",
    "        #变换x的形状以适配全连接层的输入\n",
    "        x=x.view(-1,16*5*5)\n",
    "        x=F.relu(self.fc1(x))\n",
    "        x=F.relu(self.fc2(x))\n",
    "        x=self.fc3(x)\n",
    "\n",
    "        return x\n",
    "#input[1, 1, 32, 32]->conv1poll[1, 6, 14, 14]->conv2pool[1, 16, 5, 5]->view[1, 400]->linear1relu[1, 120]->linear2relu[1,84]->linear3relu[10]\n",
    "net=Net()\n",
    "print(net)\n",
    "\n",
    "import torch.optim as optim\n",
    "\n",
    "#定义损失函数，选用交叉熵损失函数\n",
    "criterion=nn.CrossEntropyLoss()\n",
    "#定义优化器，选用随机梯度下降的优化器\n",
    "#optimizer=optim.SGD(net.parameters(),lr=0.001,momentum=0.9)\n",
    "\n",
    "\n",
    "#改为选用adam优化算法\n",
    "# define the Adam optimizer\n",
    "optimizer = optim.Adam(net.parameters(), lr=0.001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "203170ea",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 2000] loss:1.874\n",
      "[1, 4000] loss:1.610\n",
      "[1, 6000] loss:1.528\n",
      "[1, 8000] loss:1.472\n",
      "[1,10000] loss:1.446\n",
      "[1,12000] loss:1.377\n",
      "[2, 2000] loss:1.324\n",
      "[2, 4000] loss:1.302\n",
      "[2, 6000] loss:1.319\n",
      "[2, 8000] loss:1.295\n",
      "[2,10000] loss:1.281\n",
      "[2,12000] loss:1.275\n",
      "[3, 2000] loss:1.187\n",
      "[3, 4000] loss:1.192\n",
      "[3, 6000] loss:1.209\n",
      "[3, 8000] loss:1.203\n",
      "[3,10000] loss:1.188\n",
      "[3,12000] loss:1.181\n",
      "[4, 2000] loss:1.097\n",
      "[4, 4000] loss:1.127\n",
      "[4, 6000] loss:1.135\n",
      "[4, 8000] loss:1.124\n",
      "[4,10000] loss:1.127\n",
      "[4,12000] loss:1.130\n",
      "[5, 2000] loss:1.068\n",
      "[5, 4000] loss:1.065\n",
      "[5, 6000] loss:1.064\n",
      "[5, 8000] loss:1.076\n",
      "[5,10000] loss:1.070\n",
      "[5,12000] loss:1.099\n",
      "[6, 2000] loss:1.009\n",
      "[6, 4000] loss:1.023\n",
      "[6, 6000] loss:1.040\n",
      "[6, 8000] loss:1.060\n",
      "[6,10000] loss:1.039\n",
      "[6,12000] loss:1.050\n",
      "[7, 2000] loss:0.974\n",
      "[7, 4000] loss:0.973\n",
      "[7, 6000] loss:0.981\n",
      "[7, 8000] loss:1.027\n",
      "[7,10000] loss:1.015\n",
      "[7,12000] loss:1.022\n",
      "[8, 2000] loss:0.918\n",
      "[8, 4000] loss:0.960\n",
      "[8, 6000] loss:0.952\n",
      "[8, 8000] loss:0.984\n",
      "[8,10000] loss:0.998\n",
      "[8,12000] loss:1.005\n",
      "[9, 2000] loss:0.894\n",
      "[9, 4000] loss:0.956\n",
      "[9, 6000] loss:0.953\n",
      "[9, 8000] loss:0.950\n",
      "[9,10000] loss:0.970\n",
      "[9,12000] loss:0.962\n",
      "[10, 2000] loss:0.883\n",
      "[10, 4000] loss:0.909\n",
      "[10, 6000] loss:0.919\n",
      "[10, 8000] loss:0.954\n",
      "[10,10000] loss:0.943\n",
      "[10,12000] loss:0.946\n",
      "Finished training\n"
     ]
    }
   ],
   "source": [
    "#编写训练集代码\n",
    "for epoch in range(10):\n",
    "    running_loss=0.0\n",
    "    #按批次迭代训练模型\n",
    "    for i,data in enumerate(trainloader,0):\n",
    "        #从data中去除含有输入图像的张量inputs，标签张量labels\n",
    "        inputs,labels=data\n",
    "        #第一步梯度清零\n",
    "        optimizer.zero_grad()\n",
    "        #第二步将输入图像进入网络中，得到输出张量\n",
    "        outputs=net(inputs)\n",
    "        #计算损失值\n",
    "        loss=criterion(outputs,labels)\n",
    "        #进行反向传播和梯度更新\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "        #打印训练的信息\n",
    "        running_loss+=loss.item()\n",
    "        if (i+1)%2000==0:\n",
    "            print('[%d,%5d] loss:%.3f'%(epoch+1,i+1,running_loss/2000))\n",
    "            running_loss=0.0\n",
    "\n",
    "print('Finished training')\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5fbb34a4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GroundTruth   cat ship shipplane\n"
     ]
    },
    {
     "ename": "TypeError",
     "evalue": "%d format: a number is required, not str",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_15676\\650381273.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     25\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     26\u001b[0m \u001b[1;31m#打印预测标签\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 27\u001b[1;33m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'Predicted:'\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;34m''\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mjoin\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'%5d'\u001b[0m\u001b[1;33m%\u001b[0m \u001b[0mclasses\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mpredicted\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mj\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mj\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m4\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     28\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_15676\\650381273.py\u001b[0m in \u001b[0;36m<genexpr>\u001b[1;34m(.0)\u001b[0m\n\u001b[0;32m     25\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     26\u001b[0m \u001b[1;31m#打印预测标签\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 27\u001b[1;33m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'Predicted:'\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;34m''\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mjoin\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'%5d'\u001b[0m\u001b[1;33m%\u001b[0m \u001b[0mclasses\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mpredicted\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mj\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mj\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m4\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     28\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mTypeError\u001b[0m: %d format: a number is required, not str"
     ]
    }
   ],
   "source": [
    "#设定模型的保存位置\n",
    "PATH='./cifar_net.pth'\n",
    "#保存模型的状态字典\n",
    "torch.save(net.state_dict(),PATH)#***********************************************\n",
    "\n",
    "\n",
    "#在测试集中取出一个批次的数据，做图像和标签的展示\n",
    "dataiter=iter(testloader)\n",
    "images,labels=dataiter.next()\n",
    "\n",
    "#打印原始图片\n",
    "imshow(torchvision.utils.make_grid(images))\n",
    "#打印真实的标签\n",
    "print('GroundTruth',''.join('%5s'% classes[labels[j]] for j in range(4)))\n",
    "\n",
    "\n",
    "#加载模型参数，在测试阶段\n",
    "net.load_state_dict(torch.load(PATH))#*****************************************\n",
    "\n",
    "#利用模型对图片进行预测\n",
    "outputs=net(images)\n",
    "\n",
    "#模型有10个类别的输出，选取其中概率最大的那个类别作为预测值\n",
    "_,predicted=torch.max(outputs,1)\n",
    "\n",
    "#打印预测标签\n",
    "print('Predicted:',''.join('%5d'% classes[predicted[j]] for j in range(4)))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "a4a6dc84",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of the network on the 10000 test images:61 %\n",
      "Accurary of plane : 65 %\n",
      "Accurary of   car : 70 %\n",
      "Accurary of  bird : 56 %\n",
      "Accurary of   cat : 46 %\n",
      "Accurary of  deer : 51 %\n",
      "Accurary of   dog : 52 %\n",
      "Accurary of  frog : 69 %\n",
      "Accurary of horse : 68 %\n",
      "Accurary of  ship : 71 %\n",
      "Accurary of truck : 67 %\n"
     ]
    }
   ],
   "source": [
    "#在这整个测试集上测试模型的准确率\n",
    "correct=0\n",
    "total=0\n",
    "with torch.no_grad():\n",
    "    for data in testloader:\n",
    "        images,labels=data\n",
    "        outputs=net(images)\n",
    "        _,predicted=torch.max(outputs.data,1)\n",
    "        total+=labels.size(0)\n",
    "        correct+=(predicted==labels).sum().item()\n",
    "\n",
    "print('Accuracy of the network on the 10000 test images:%d %%'%(100*correct/total))\n",
    "\n",
    "\n",
    "#分别测试不同类别的模型准确率\n",
    "class_correct=list(0. for i in range(10))\n",
    "class_total=list(0. for i in range(10))\n",
    "\n",
    "with torch.no_grad():\n",
    "    for data in testloader:\n",
    "        images,labels=data\n",
    "        outputs=net(images)\n",
    "        _,predicted=torch.max(outputs,1)\n",
    "        c=(predicted==labels).squeeze()\n",
    "        for i in range(4):\n",
    "            label=labels[i]\n",
    "            class_correct[label]+=c[i].item()\n",
    "            class_total[label]+=1\n",
    "\n",
    "#打印不同类别的准确率\n",
    "for i in range(10):\n",
    "    print('Accurary of %5s : %2d %%'% (classes[i],100*class_correct[i]/class_total[i]))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "69d714ec",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda:0\n"
     ]
    }
   ],
   "source": [
    "#在GPU上训练模型\n",
    "device=torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "print(device)\n",
    "\n",
    "#将模型转移到GPU上\n",
    "net.to(device)\n",
    "\n",
    "#将输入的图片张量和标签张量转移到GPU上\n",
    "inputs,labels=data[0].to(device),data[1].to(device)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "321a8954",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}