第13周实验-基于卷积神经网络的猫狗识别
基于卷积神经网络的猫狗识别
- 任务需求
- 环境配置
- 识别实例
- 结果
- 参考
任务需求
按照 https://github.com/fchollet/deep-learning-with-python-notebooks/blob/master/5.2-using-convnets-with-small-datasets.ipynb,利用TensorFlow和Keras，自己搭建卷积神经网络完成狗猫数据集的分类实验；将关键步骤用汉语注释出来。
解释什么是overfit（过拟合）?什么是数据增强？
如果单独只做数据增强，精确率提高了多少？
然后再添加的dropout层，是什么实际效果？
环境配置
- 安装Anaconda
- Pytorch
代码
pip install -i https://pypi.tuna.tsinghua.edu.cn/simple torch
pip install -i https://pypi.tuna.tsinghua.edu.cn/simple torchvision
构建数据集
构建数据集地址：https://www.kaggle.com/lizhensheng/-2000
识别实例
- 导入库
# 导入库
import torch.nn.functional as F
import torch.optim as optim
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
import torch.utils.data.distributed
import torchvision.transforms as transforms
import torchvision.datasets as datasets
# 设置超参数
BATCH_SIZE = 20
EPOCHS = 10
DEVICE = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
- 预处理数据
# 数据预处理
transform = transforms.Compose([
transforms.Resize(100),
transforms.RandomVerticalFlip(),
transforms.RandomCrop(50),
transforms.RandomResizedCrop(150),
transforms.ColorJitter(brightness=0.5, contrast=0.5, hue=0.5),
transforms.ToTensor(),
transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])
])
- 读取需要的数据
# 读取数据
dataset_train = datasets.ImageFolder('data/train', transform)
print(dataset_train.imgs)
# 对应文件夹的label
print(dataset_train.class_to_idx)
dataset_test = datasets.ImageFolder('data/val', transform)
# 对应文件夹的label
print(dataset_test.class_to_idx)
# 导入数据
train_loader = torch.utils.data.DataLoader(dataset_train, batch_size=BATCH_SIZE, shuffle=True)
test_loader = torch.utils.data.DataLoader(dataset_test, batch_size=BATCH_SIZE, shuffle=True)
- 定义网络
# 定义网络
class ConvNet(nn.Module):
def __init__(self):
super(ConvNet, self).__init__()
self.conv1 = nn.Conv2d(3, 32, 3)
self.max_pool1 = nn.MaxPool2d(2)
self.conv2 = nn.Conv2d(32, 64, 3)
self.max_pool2 = nn.MaxPool2d(2)
self.conv3 = nn.Conv2d(64, 64, 3)
self.conv4 = nn.Conv2d(64, 64, 3)
self.max_pool3 = nn.MaxPool2d(2)
self.conv5 = nn.Conv2d(64, 128, 3)
self.conv6 = nn.Conv2d(128, 128, 3)
self.max_pool4 = nn.MaxPool2d(2)
self.fc1 = nn.Linear(4608, 512)
self.fc2 = nn.Linear(512, 1)
def forward(self, x):
in_size = x.size(0)
x = self.conv1(x)
x = F.relu(x)
x = self.max_pool1(x)
x = self.conv2(x)
x = F.relu(x)
x = self.max_pool2(x)
x = self.conv3(x)
x = F.relu(x)
x = self.conv4(x)
x = F.relu(x)
x = self.max_pool3(x)
x = self.conv5(x)
x = F.relu(x)
x = self.conv6(x)
x = F.relu(x)
x = self.max_pool4(x)
# 展开
x = x.view(in_size, -1)
x = self.fc1(x)
x = F.relu(x)
x = self.fc2(x)
x = torch.sigmoid(x)
return x
modellr = 1e-4
# 实例化模型并且移动到GPU
model = ConvNet().to(DEVICE)
# 选择简单暴力的Adam优化器，学习率调低
optimizer = optim.Adam(model.parameters(), lr=modellr)
- 机器学习率
def adjust_learning_rate(optimizer, epoch):
"""Sets the learning rate to the initial LR decayed by 10 every 30 epochs"""
modellrnew = modellr * (0.1 ** (epoch // 5))
print("lr:",modellrnew)
for param_group in optimizer.param_groups:
param_group['lr'] = modellrnew
- 训练过程
# 定义训练过程
def train(model, device, train_loader, optimizer, epoch):
model.train()
for batch_idx, (data, target) in enumerate(train_loader):
data, target = data.to(device), target.to(device).float().unsqueeze(1)
optimizer.zero_grad()
output = model(data)
# print(output)
loss = F.binary_cross_entropy(output, target)
loss.backward()
optimizer.step()
if (batch_idx + 1) % 10 == 0:
print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
epoch, (batch_idx + 1) * len(data), len(train_loader.dataset),
100. * (batch_idx + 1) / len(train_loader), loss.item()))
# 定义测试过程
def val(model, device, test_loader):
model.eval()
test_loss = 0
correct = 0
with torch.no_grad():
for data, target in test_loader:
data, target = data.to(device), target.to(device).float().unsqueeze(1)
output = model(data)
# print(output)
test_loss += F.binary_cross_entropy(output, target, reduction='mean').item() # 将一批的损失相加
pred = torch.tensor([[1] if num[0] >= 0.5 else [0] for num in output]).to(device)
correct += pred.eq(target.long()).sum().item()
print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
test_loss, correct, len(test_loader.dataset),
100. * correct / len(test_loader.dataset)))
- 开始训练
# 训练
for epoch in range(1, EPOCHS + 1):
adjust_learning_rate(optimizer, epoch)
train(model, DEVICE, train_loader, optimizer, epoch)
val(model, DEVICE, test_loader)
torch.save(model, 'model.pth')
- 训练结果
- 测试所使用的代码
from __future__ import print_function, division
from PIL import Image
from torchvision import transforms
import torch.nn.functional as F
import torch
import torch.nn as nn
import torch.nn.parallel
# 定义网络
class ConvNet(nn.Module):
def __init__(self):
super(ConvNet, self).__init__()
self.conv1 = nn.Conv2d(3, 32, 3)
self.max_pool1 = nn.MaxPool2d(2)
self.conv2 = nn.Conv2d(32, 64, 3)
self.max_pool2 = nn.MaxPool2d(2)
self.conv3 = nn.Conv2d(64, 64, 3)
self.conv4 = nn.Conv2d(64, 64, 3)
self.max_pool3 = nn.MaxPool2d(2)
self.conv5 = nn.Conv2d(64, 128, 3)
self.conv6 = nn.Conv2d(128, 128, 3)
self.max_pool4 = nn.MaxPool2d(2)
self.fc1 = nn.Linear(4608, 512)
self.fc2 = nn.Linear(512, 1)
def forward(self, x):
in_size = x.size(0)
x = self.conv1(x)
x = F.relu(x)
x = self.max_pool1(x)
x = self.conv2(x)
x = F.relu(x)
x = self.max_pool2(x)
x = self.conv3(x)
x = F.relu(x)
x = self.conv4(x)
x = F.relu(x)
x = self.max_pool3(x)
x = self.conv5(x)
x = F.relu(x)
x = self.conv6(x)
x = F.relu(x)
x = self.max_pool4(x)
# 展开
x = x.view(in_size, -1)
x = self.fc1(x)
x = F.relu(x)
x = self.fc2(x)
x = torch.sigmoid(x)
return x
# 模型存储路径
model_save_path = 'model.pth'
# Data augmentation and normalization for training
# Just normalization for validation
# 定义预训练变换
# 数据预处理
transform_test = transforms.Compose([
transforms.Resize(100),
transforms.RandomVerticalFlip(),
transforms.RandomCrop(50),
transforms.RandomResizedCrop(150),
transforms.ColorJitter(brightness=0.5, contrast=0.5, hue=0.5),
transforms.ToTensor(),
transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])
])
class_names = ['cat', 'dog']
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model = torch.load(model_save_path)
model.eval()
# print(model)
image_PIL = Image.open('dog.12.jpg')
#
image_tensor = transform_test(image_PIL)
# 以下语句等效于 image_tensor = torch.unsqueeze(image_tensor, 0)
image_tensor.unsqueeze_(0)
image_tensor = image_tensor.to(device)
out = model(image_tensor)
pred = torch.tensor([[1] if num[0] >= 0.5 else [0] for num in out]).to(device)
print(class_names[pred])
结果
参考
参考链接