位置: IT常识 - 正文
推荐整理分享基于Pytorch的MNIST手写数字识别实现(含代码+讲解)(基于Pytorch的风格转换),希望有所帮助,仅作参考,欢迎阅读内容。
文章相关热门搜索词:基于Pytorch的LSTM模型对股价的分析与预测,基于pytorch的图像分类算法,基于Pytorch的垃圾分类毕业论文,基于Pytorch的结构化剪枝,基于Pytorch的结构化剪枝,基于Pytorch的结构化剪枝,基于Pytorch的LSTM模型对股价的分析与预测,基于Pytorch的车道线检测,内容如对您有帮助,希望把文章链接给更多的朋友!
说明:本人也是一个萌新,也在学习中,有代码里也有不完善的地方。如果有错误/讲解不清的地方请多多指出
本文代码链接:GitHub - Michael-OvO/mnist: mnist_trained_model with torch
明确任务目标:使用pytorch作为框架使用mnist数据集训练一个手写数字的识别
换句话说:输入为
输出: 0
比较简单直观
1. 环境搭建需要安装Pytorch, 具体过程因系统而异,这里也就不多赘述了
具体教程可以参考这个视频 (这个系列的P1是环境配置)
PyTorch深度学习快速入门教程(绝对通俗易懂!)【小土堆】_哔哩哔哩_bilibili【已完结!!!已完结!!!2021年5月31日已完结】本系列教程,将带你用全新的思路,快速入门PyTorch。独创的学习思路,仅此一家。个人公众号:我是土堆各种资料,请自取。代码:https://github.com/xiaotudui/PyTorch-Tutorial蚂蚁蜜蜂/练手数据集:链接: https://pan.baidu.com/s/1jZoTmoFzaTLWh4lKBHVbEA 密码https://www.bilibili.com/video/BV1hE411t7RN?share_source=copy_web
2. 基本导入import torchimport torchvisionfrom torch.utils.data import DataLoaderimport torch.nn as nnimport torch.optim as optimfrom torch.utils.tensorboard import SummaryWriterimport timeimport matplotlib.pyplot as pltimport randomfrom numpy import argmax不多解释,导入各种需要的包
3. 基本参数定义#Basic Params-----------------------------epoch = 1learning_rate = 0.01batch_size_train = 64batch_size_test = 1000gpu = torch.cuda.is_available()momentum = 0.5epoch是整体进行几批训练
learning rate 为学习率
随后是每批训练数据大小和测试数据大小
gpu是一个布尔值,方便没有显卡的同学可以不用cuda加速,但是有显卡的同学可以优先使用cuda
momentum 是动量,避免找不到局部最优解的尴尬情况
这些都是比较基本的网络参数
4. 数据加载使用Dataloader加载数据,如果是第一次运行将会从网上下载数据
如果下载一直不行的话也可以从官方直接下载并放入./data目录即可
MNIST handwritten digit database, Yann LeCun, Corinna Cortes and Chris Burges
(有4个包都需要下载)
#Load Data-------------------------------train_loader = DataLoader(torchvision.datasets.MNIST('./data/', train=True, download=True, transform=torchvision.transforms.Compose([ torchvision.transforms.ToTensor(), torchvision.transforms.Normalize( (0.1307,), (0.3081,)) ])), batch_size=batch_size_train, shuffle=True)test_loader = DataLoader(torchvision.datasets.MNIST('./data/', train=False, download=True, transform=torchvision.transforms.Compose([ torchvision.transforms.ToTensor(), torchvision.transforms.Normalize( (0.1307,), (0.3081,)) ])), batch_size=batch_size_test, shuffle=True)train_data_size = len(train_loader)test_data_size = len(test_loader)5. 网络定义接下来是重中之重
网络的定义
这边的网络结构参考了这张图:
有了结构图,代码就很好写了, 直接对着图敲出来就好了
非常建议使用sequential直接写网络结构,会方便很多
#Define Model----------------------------class Net(nn.Module): def __init__(self): super(Net,self).__init__() self.model = nn.Sequential( nn.Conv2d(in_channels=1, out_channels=32, kernel_size=3, stride=1, padding=1), nn.ReLU(), nn.MaxPool2d(kernel_size=2, stride=2), nn.Conv2d(in_channels=32, out_channels=64, kernel_size=3, stride=1, padding=1), nn.ReLU(), nn.MaxPool2d(kernel_size=2, stride=2), nn.Flatten(), nn.Linear(in_features=3136, out_features=128), nn.Linear(in_features=128, out_features=10), ) def forward(self, x): return self.model(x)if gpu: net = Net().cuda()else: net = Net()6.损失函数和优化器交叉熵和SGD(随机梯度下降)
另外为了方便记录训练情况可以使用TensorBoard的Summary Writer
#Define Loss and Optimizer----------------if gpu: loss_fn = nn.CrossEntropyLoss().cuda()else: loss_fn = nn.CrossEntropyLoss()optimizer = optim.SGD(net.parameters(), lr=learning_rate, momentum=0.9)#Define Tensorboard-------------------writer = SummaryWriter(log_dir='logs/{}'.format(time.strftime('%Y%m%d-%H%M%S')))7. 开始训练#Train---------------------------------total_train_step = 0def train(epoch): global total_train_step total_train_step = 0 for data in train_loader: imgs,targets = data if gpu: imgs,targets = imgs.cuda(),targets.cuda() optimizer.zero_grad() outputs = net(imgs) loss = loss_fn(outputs,targets) loss.backward() optimizer.step() if total_train_step % 200 == 0: print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format( epoch, total_train_step, train_data_size, 100. * total_train_step / train_data_size, loss.item())) writer.add_scalar('loss', loss.item(), total_train_step) total_train_step += 1#Test---------------------------------def test(): correct = 0 total = 0 with torch.no_grad(): for data in test_loader: imgs,targets = data if gpu: imgs,targets = imgs.cuda(),targets.cuda() outputs = net(imgs) _,predicted = torch.max(outputs.data,1) total += targets.size(0) correct += (predicted == targets).sum().item() print('Test Accuracy: {}/{} ({:.0f}%)'.format(correct, total, 100.*correct/total)) return correct/total#Run----------------------------------for i in range(1,epoch+1): print("-----------------Epoch: {}-----------------".format(i)) train(i) test() writer.add_scalar('test_accuracy', test(), total_train_step) #save model torch.save(net,'model/mnist_model.pth') print('Saved model')writer.close()注意这里必须要先在同一文件夹下创建一个叫做model的文件夹!!!不然模型目录将找不到地方保存!!!会报错!!!
Train函数作为训练,Test函数作为测试
注意每次训练需要梯度清零
模型测试时要写with torch.no_grad()
运行的过程如果有GPU加速会很快,运行结果应该如下
正确率也还算是可以,一个epoch就能跑到98,如果不满意或者想调epoch次数可以在basic params区域直接进行修改
8. 模型验证和结果展示小细节很多
首先是抽取样本的时候需要考虑转cuda的问题
其次如果直接将样本扔到网络里dimension不对,需要reshape
需要对结果进行argmax处理,因为结果是一个向量(有10个features,分别代表每个数字的概率),argmax会找到最大概率并输出模型的预测结果
使用matplotlib画图
#Evaluate---------------------------------model = torch.load("./model/mnist_model.pth")model.eval()print(model)fig = plt.figure(figsize=(20,20))for i in range(20): #随机抽取20个样本 index = random.randint(0,test_data_size) data = test_loader.dataset[index] #注意Cuda问题 if gpu: img = data[0].cuda() else: img = data[0] #维度不对必须要reshape img = torch.reshape(img,(1,1,28,28)) with torch.no_grad(): output = model(img) #plot the image and the predicted number fig.add_subplot(4,5,i+1) #一定要取Argmax!!! plt.title(argmax(output.data.cpu().numpy())) plt.imshow(data[0].numpy().squeeze(),cmap='gray')plt.show()运行结果如下:
效果还是很不错的
至此我们就完成了一整个模型训练,保存,导入,验证的基本流程。
完整代码import torchimport torchvisionfrom torch.utils.data import DataLoaderimport torch.nn as nnimport torch.optim as optimfrom torch.utils.tensorboard import SummaryWriterimport timeimport matplotlib.pyplot as pltimport randomfrom numpy import argmax#Basic Params-----------------------------epoch = 1learning_rate = 0.01batch_size_train = 64batch_size_test = 1000gpu = torch.cuda.is_available()momentum = 0.5#Load Data-------------------------------train_loader = DataLoader(torchvision.datasets.MNIST('./data/', train=True, download=True, transform=torchvision.transforms.Compose([ torchvision.transforms.ToTensor(), torchvision.transforms.Normalize( (0.1307,), (0.3081,)) ])), batch_size=batch_size_train, shuffle=True)test_loader = DataLoader(torchvision.datasets.MNIST('./data/', train=False, download=True, transform=torchvision.transforms.Compose([ torchvision.transforms.ToTensor(), torchvision.transforms.Normalize( (0.1307,), (0.3081,)) ])), batch_size=batch_size_test, shuffle=True)train_data_size = len(train_loader)test_data_size = len(test_loader)#Define Model----------------------------class Net(nn.Module): def __init__(self): super(Net,self).__init__() self.model = nn.Sequential( nn.Conv2d(in_channels=1, out_channels=32, kernel_size=3, stride=1, padding=1), nn.ReLU(), nn.MaxPool2d(kernel_size=2, stride=2), nn.Conv2d(in_channels=32, out_channels=64, kernel_size=3, stride=1, padding=1), nn.ReLU(), nn.MaxPool2d(kernel_size=2, stride=2), nn.Flatten(), nn.Linear(in_features=3136, out_features=128), nn.Linear(in_features=128, out_features=10), ) def forward(self, x): return self.model(x)if gpu: net = Net().cuda()else: net = Net()#Define Loss and Optimizer----------------if gpu: loss_fn = nn.CrossEntropyLoss().cuda()else: loss_fn = nn.CrossEntropyLoss()optimizer = optim.SGD(net.parameters(), lr=learning_rate, momentum=0.9)#Define Tensorboard-------------------writer = SummaryWriter(log_dir='logs/{}'.format(time.strftime('%Y%m%d-%H%M%S')))#Train---------------------------------total_train_step = 0def train(epoch): global total_train_step total_train_step = 0 for data in train_loader: imgs,targets = data if gpu: imgs,targets = imgs.cuda(),targets.cuda() optimizer.zero_grad() outputs = net(imgs) loss = loss_fn(outputs,targets) loss.backward() optimizer.step() if total_train_step % 200 == 0: print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format( epoch, total_train_step, train_data_size, 100. * total_train_step / train_data_size, loss.item())) writer.add_scalar('loss', loss.item(), total_train_step) total_train_step += 1#Test---------------------------------def test(): correct = 0 total = 0 with torch.no_grad(): for data in test_loader: imgs,targets = data if gpu: imgs,targets = imgs.cuda(),targets.cuda() outputs = net(imgs) _,predicted = torch.max(outputs.data,1) total += targets.size(0) correct += (predicted == targets).sum().item() print('Test Accuracy: {}/{} ({:.0f}%)'.format(correct, total, 100.*correct/total)) return correct/total#Run----------------------------------for i in range(1,epoch+1): print("-----------------Epoch: {}-----------------".format(i)) train(i) test() writer.add_scalar('test_accuracy', test(), total_train_step) #save model torch.save(net,'model/mnist_model.pth') print('Saved model')writer.close()#Evaluate---------------------------------model = torch.load("./model/mnist_model.pth")model.eval()print(model)fig = plt.figure(figsize=(20,20))for i in range(20): #random number index = random.randint(0,test_data_size) data = test_loader.dataset[index] if gpu: img = data[0].cuda() else: img = data[0] img = torch.reshape(img,(1,1,28,28)) with torch.no_grad(): output = model(img) #plot the image and the predicted number fig.add_subplot(4,5,i+1) plt.title(argmax(output.data.cpu().numpy())) plt.imshow(data[0].numpy().squeeze(),cmap='gray')plt.show()上一篇:【yolov6系列一】深度解析网络架构(yolov5官方)
下一篇:【Spring】IOC,你真的懂了吗?(spring ioc di aop)
友情链接: 武汉网站建设