vgg16_main.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "import cv2 as cv\n",
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn.functional as F\n",
    "import torch.utils.data as data\n",
    "import torchvision.transforms as transforms\n",
    "from matplotlib import pyplot as plt\n",
    "from PIL import Image\n",
    "from torch import nn, optim\n",
    "from torch.utils.data import DataLoader\n",
    "from my_models import *\n",
    "from cat_dog_dataset import my_Dataset\n",
    "global PATH, IMG_SIZE, BATCH_SIZE, DEVICE\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def train(epoch, model, train_dataloader, criterion, optimizer):\n",
    "    '''模型训练函数'''\n",
    "    global print_loss\n",
    "    # 调整为训练模式\n",
    "    model.train()\n",
    "    losses = []\n",
    "    plot_loss = 0\n",
    "    total_num = len(train_dataloader.dataset)\n",
    "    print_thre = total_num//(BATCH_SIZE*5)\n",
    "    for i, (img, label) in enumerate(train_dataloader):\n",
    "        img, label = img.to(DEVICE), label.to(DEVICE)\n",
    "        out = model(img)\n",
    "        # 计算损失\n",
    "        loss = criterion(out, label.squeeze())\n",
    "        losses.append(loss.item())\n",
    "        # 误差反向传播\n",
    "        loss.backward()\n",
    "        # 优化采用设定的优化方法对网络中的各个参数进行调整\n",
    "        optimizer.step()\n",
    "        # 清除优化器中的梯度以便下一次计算\n",
    "        optimizer.zero_grad()\n",
    "        if (i+1) % print_thre == 0:\n",
    "            plot_loss += np.mean(losses)\n",
    "            print('TRAINING: Epoch:{} [{} / {} ({:.2f}%)] Loss:{}'.format(\n",
    "                epoch, (i+1)*BATCH_SIZE, total_num, (i+1)*BATCH_SIZE/total_num*100, plot_loss))\n",
    "            losses.clear()\n",
    "    print_loss.append(plot_loss/5)\n",
    "\n",
    "\n",
    "def test(model, test_dataloader, criterion):\n",
    "    '''模型验证函数'''\n",
    "    model.eval()\n",
    "    cnt = 0\n",
    "    losses = []\n",
    "    total_num = len(test_dataloader.dataset)\n",
    "    for i, (img, label) in enumerate(test_dataloader):\n",
    "        img, label = img.to(DEVICE), label.to(DEVICE)\n",
    "        out = model(img)\n",
    "        # 计算损失\n",
    "        loss = criterion(out, label.squeeze())\n",
    "        losses.append(loss.item())\n",
    "        y = out.argmax(1, keepdim=True)\n",
    "        num = (y-label).abs().sum()\n",
    "        cnt += num\n",
    "    acc = 1-cnt/total_num\n",
    "    print('\\nTESTING: Loss:{},Accuracy:[{} / {} ({:.6f}%)]\\n'.format(np.mean(losses), total_num-cnt, total_num,\n",
    "          acc*100))\n",
    "    return acc.item()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "using device cuda\n"
     ]
    }
   ],
   "source": [
    "global PATH, IMG_SIZE, BATCH_SIZE, DEVICE\n",
    "'''定义相关超参数'''\n",
    "# 训练集路径\n",
    "PATH = \"data/train\"\n",
    "# 裁剪图像大小\n",
    "IMG_SIZE = 256\n",
    "BATCH_SIZE = 25\n",
    "EPOCHS = 5\n",
    "DEVICE = torch.device(\n",
    "    \"cuda\") if torch.cuda.is_available() else torch.device(\"cpu\")\n",
    "WORKERS = 0\n",
    "# 训练数据比例\n",
    "TRAIN_RATE = 0.7\n",
    "\n",
    "print(\"using device {}\".format(DEVICE))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# 数据集分割，采用固定的seed\n",
    "img_data = my_Dataset(\"train\", PATH, IMG_SIZE)\n",
    "IMG_NUM = len(img_data)\n",
    "TRAIN_NUM = int(IMG_NUM*TRAIN_RATE)\n",
    "train_data, vali_data = data.random_split(\n",
    "    img_data, [TRAIN_NUM, IMG_NUM-TRAIN_NUM], torch.Generator().manual_seed(1))\n",
    "train_dataloader = DataLoader(train_data, BATCH_SIZE,\n",
    "                                shuffle=True, num_workers=WORKERS, drop_last=True)\n",
    "vali_dataloader = DataLoader(vali_data, BATCH_SIZE,\n",
    "                                shuffle=True, num_workers=WORKERS, drop_last=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "using model: VGG\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# 定义网络模型\n",
    "# model = get_resnet18_model(False, True).to(DEVICE)\n",
    "model = get_vgg16_model(True, True).to(DEVICE)\n",
    "print(\"using model:\", model._get_name())\n",
    "# 实例化一个优化器，优化方式为adam方法\n",
    "optimizer = optim.Adam(model.parameters())\n",
    "# 定义损失函数为交叉熵函数\n",
    "criterion = nn.CrossEntropyLoss()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRAINING: Epoch:0 [3500 / 17500 (20.00%)] Loss:0.3450650210891451\n",
      "TRAINING: Epoch:0 [7000 / 17500 (40.00%)] Loss:0.686212545633316\n",
      "TRAINING: Epoch:0 [10500 / 17500 (60.00%)] Loss:1.0322584654603686\n",
      "TRAINING: Epoch:0 [14000 / 17500 (80.00%)] Loss:1.3720338900174414\n",
      "TRAINING: Epoch:0 [17500 / 17500 (100.00%)] Loss:1.7174275293946266\n",
      "\n",
      "TESTING: Loss:0.3427267023921013,Accuracy:[7272 / 7500 (96.959999%)]\n",
      "\n",
      "TRAINING: Epoch:1 [3500 / 17500 (20.00%)] Loss:0.3468616443020957\n",
      "TRAINING: Epoch:1 [7000 / 17500 (40.00%)] Loss:0.7007708257862499\n",
      "TRAINING: Epoch:1 [10500 / 17500 (60.00%)] Loss:1.0468020839350562\n",
      "TRAINING: Epoch:1 [14000 / 17500 (80.00%)] Loss:1.3889035061001775\n",
      "TRAINING: Epoch:1 [17500 / 17500 (100.00%)] Loss:1.7306148878165653\n",
      "\n",
      "TESTING: Loss:0.3435409663120906,Accuracy:[7268 / 7500 (96.906670%)]\n",
      "\n",
      "TRAINING: Epoch:2 [3500 / 17500 (20.00%)] Loss:0.342246484543596\n",
      "TRAINING: Epoch:2 [7000 / 17500 (40.00%)] Loss:0.6852805535708155\n",
      "TRAINING: Epoch:2 [10500 / 17500 (60.00%)] Loss:1.0326016962528228\n",
      "TRAINING: Epoch:2 [14000 / 17500 (80.00%)] Loss:1.3749851850526673\n",
      "TRAINING: Epoch:2 [17500 / 17500 (100.00%)] Loss:1.7215096833450454\n",
      "\n",
      "TESTING: Loss:0.3440323449174563,Accuracy:[7263 / 7500 (96.839996%)]\n",
      "\n",
      "TRAINING: Epoch:3 [3500 / 17500 (20.00%)] Loss:0.3456301529492651\n",
      "TRAINING: Epoch:3 [7000 / 17500 (40.00%)] Loss:0.6919617867895536\n",
      "TRAINING: Epoch:3 [10500 / 17500 (60.00%)] Loss:1.0414420442921775\n",
      "TRAINING: Epoch:3 [14000 / 17500 (80.00%)] Loss:1.3875373908451625\n",
      "TRAINING: Epoch:3 [17500 / 17500 (100.00%)] Loss:1.7295707960213933\n",
      "\n",
      "TESTING: Loss:0.34256623635689415,Accuracy:[7274 / 7500 (96.986671%)]\n",
      "\n",
      "TRAINING: Epoch:4 [3500 / 17500 (20.00%)] Loss:0.34466694401843206\n",
      "TRAINING: Epoch:4 [7000 / 17500 (40.00%)] Loss:0.6905048815267427\n",
      "TRAINING: Epoch:4 [10500 / 17500 (60.00%)] Loss:1.0375139481254987\n",
      "TRAINING: Epoch:4 [14000 / 17500 (80.00%)] Loss:1.3800737444843565\n",
      "TRAINING: Epoch:4 [17500 / 17500 (100.00%)] Loss:1.7224809876510077\n",
      "\n",
      "TESTING: Loss:0.3438378084699313,Accuracy:[7262 / 7500 (96.826668%)]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print_loss = []\n",
    "accs = []\n",
    "for epoch in range(EPOCHS):\n",
    "    train(epoch, model, train_dataloader, criterion, optimizer)\n",
    "    if (epoch+1) % 1 == 0:\n",
    "        acc = test(model, vali_dataloader, criterion)\n",
    "        accs.append(acc)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.save(\n",
    "    model.classifier._modules[\"6\"].state_dict(), model._get_name()+\"_state.pth\")\n",
    "print(model._get_name()+\"模型参数保存成功\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(1)\n",
    "plt.plot(range(EPOCHS), print_loss,   linestyle='-.', linewidth=2,\n",
    "         marker='o', markersize=8, markeredgecolor='k', markerfacecolor='red')\n",
    "plt.xlabel(\"EPOCHS\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.xticks(range(EPOCHS))\n",
    "plt.title(\"VGG16 Loss下降曲线\")\n",
    "plt.savefig(\"Loss_vgg16.jpg\")\n",
    "\n",
    "plt.figure(2)\n",
    "plt.plot(range(EPOCHS), accs,linestyle='-.', linewidth=2,\n",
    "         marker='o', markersize=8, markeredgecolor='k', markerfacecolor='red')\n",
    "plt.xlabel(\"EPOCHS\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.xticks(range(EPOCHS))\n",
    "plt.title(\"vgg16 准确率变化情况\")\n",
    "plt.savefig(\"Accuracy_vgg16.jpg\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "模型参数保存成功\n"
     ]
    }
   ],
   "source": [
    "torch.save(model.state_dict(), \"vgg_model_state.pth\")\n",
    "print(\"模型参数保存成功\")\n"
   ]
  }
 ],
 "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": "3.9.1 (tags/v3.9.1:1e5d33e, Dec  7 2020, 17:08:21) [MSC v.1927 64 bit (AMD64)]"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "11938c6bc6919ae2720b4d5011047913343b08a43b18698fd82dedb0d4417594"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}