基于经典网络架构训练图像分类模型的学习

记录学习图像分类神经网络的学习笔记

第一步：导入所需要的库

import os
import matplotlib.pyplot as plt
%matplotlib inline
import numpy as np
import torch
from torch import nn
import torch.optim as optim
import torchvision
#pip install torchvision
from torchvision import transforms, models, datasets
#https://pytorch.org/docs/stable/torchvision/index.html
import imageio
import time
import warnings
warnings.filterwarnings("ignore")
import random
import sys
import copy
import json
from PIL import Image

第二步：数据处理

1.定义数据的文件夹

data_dir = './flower_data/'
train_dir = data_dir + '/train'
valid_dir = data_dir + '/valid'

数据分为训练集和验证集两部分，以文件夹的方式存放，文件夹的名字就是花朵的类别，每个文件夹有102个类别。具体如下图所示：

 

 2.数据增强

用的是transforms函数中的操作，包括裁剪，旋转啊等等。

data_transforms = {
    'train': 
        transforms.Compose([
        transforms.Resize([96, 96]),
        transforms.RandomRotation(45),#随机旋转，-45到45度之间随机选
        transforms.CenterCrop(64),#从中心开始裁剪
        transforms.RandomHorizontalFlip(p=0.5),#随机水平翻转 选择一个概率概率
        transforms.RandomVerticalFlip(p=0.5),#随机垂直翻转
        transforms.ColorJitter(brightness=0.2, contrast=0.1, saturation=0.1, hue=0.1),#参数1为亮度，参数2为对比度，参数3为饱和度，参数4为色相
        transforms.RandomGrayscale(p=0.025),#概率转换成灰度率，3通道就是R=G=B
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])#均值，标准差
    ]),
    'valid': 
        transforms.Compose([
        transforms.Resize([64, 64]),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ]),
}

3.定义dataloader

batch_size = 128

image_datasets = {x: datasets.ImageFolder(os.path.join(data_dir, x), data_transforms[x]) for x in ['train', 'valid']}
dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=batch_size, shuffle=True) for x in ['train', 'valid']}
dataset_sizes = {x: len(image_datasets[x]) for x in ['train', 'valid']}
class_names = image_datasets['train'].classes

 第三步：定义模型

当feature_extract=True时，代表不更新参数，feature_extract=False时，代表更新参数

model_name = 'resnet'  #可选的比较多 ['resnet', 'alexnet', 'vgg', 'squeezenet', 'densenet', 'inception']
#是否用人家训练好的特征来做
feature_extract = True #都用人家特征，咱先不更新
# 是否用GPU训练
train_on_gpu = torch.cuda.is_available()

if not train_on_gpu:
    print('CUDA is not available.  Training on CPU ...')
else:
    print('CUDA is available!  Training on GPU ...')
    
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")

1.定义一个函数来确定是否更新参数

def set_parameter_requires_grad(model, feature_extracting):
    if feature_extracting:
        for param in model.parameters():
            param.requires_grad = False

2.i定义模型，采用torhvision.models中的resnet模块，常见的有18, 34, 50, 101, 152层，这里出于对下载速度快的考虑，采用的是18层的resnet

model_ft = models.resnet18()#18层的能快点，条件好点的也可以选152
model_ft

3.更改最后一层的FC的参数，使之符合自己训练模型的大小

def initialize_model(model_name, num_classes, feature_extract, use_pretrained=True):
    
    model_ft = models.resnet18(pretrained=use_pretrained)
    set_parameter_requires_grad(model_ft, feature_extract)
    
    num_ftrs = model_ft.fc.in_features
    model_ft.fc = nn.Linear(num_ftrs, 102)#类别数自己根据自己任务来
                            
    input_size = 64#输入大小根据自己配置来

    return model_ft, input_size

第四步：优化器和损失函数设置

学习率没迭代10次衰减十分之一

# 优化器设置
optimizer_ft = optim.Adam(params_to_update, lr=1e-2)#要训练啥参数，你来定
scheduler = optim.lr_scheduler.StepLR(optimizer_ft, step_size=10, gamma=0.1)#学习率每7个epoch衰减成原来的1/10
criterion = nn.CrossEntropyLoss()

第五步：训练模型

分为训练和验证两个部分，其中训练包括梯度下降和反向传播两部分，而验证的时候没有。

训练——>验证——>保存模型

def train_model(model, dataloaders, criterion, optimizer, num_epochs=25,filename='best.pt'):
    #咱们要算时间的
    since = time.time()
    #也要记录最好的那一次
    best_acc = 0
    #模型也得放到你的CPU或者GPU
    model.to(device)
    #训练过程中打印一堆损失和指标
    val_acc_history = []
    train_acc_history = []
    train_losses = []
    valid_losses = []
    #学习率
    LRs = [optimizer.param_groups[0]['lr']]
    #最好的那次模型，后续会变的，先初始化
    best_model_wts = copy.deepcopy(model.state_dict())
    #一个个epoch来遍历
    for epoch in range(num_epochs):
        print('Epoch {}/{}'.format(epoch, num_epochs - 1))
        print('-' * 10)

        # 训练和验证
        for phase in ['train', 'valid']:
            if phase == 'train':
                model.train()  # 训练
            else:
                model.eval()   # 验证

            running_loss = 0.0
            running_corrects = 0

            # 把数据都取个遍
            for inputs, labels in dataloaders[phase]:
                inputs = inputs.to(device)#放到你的CPU或GPU
                labels = labels.to(device)

                # 清零
                optimizer.zero_grad()
                # 只有训练的时候计算和更新梯度
                outputs = model(inputs)
                loss = criterion(outputs, labels)
                _, preds = torch.max(outputs, 1)
                # 训练阶段更新权重
                if phase == 'train':
                    loss.backward()
                    optimizer.step()

                # 计算损失
                running_loss += loss.item() * inputs.size(0)#0表示batch那个维度
                running_corrects += torch.sum(preds == labels.data)#预测结果最大的和真实值是否一致
                
            
            
            epoch_loss = running_loss / len(dataloaders[phase].dataset)#算平均
            epoch_acc = running_corrects.double() / len(dataloaders[phase].dataset)
            
            time_elapsed = time.time() - since#一个epoch我浪费了多少时间
            print('Time elapsed {:.0f}m {:.0f}s'.format(time_elapsed // 60, time_elapsed % 60))
            print('{} Loss: {:.4f} Acc: {:.4f}'.format(phase, epoch_loss, epoch_acc))
            

            # 得到最好那次的模型
            if phase == 'valid' and epoch_acc > best_acc:
                best_acc = epoch_acc
                best_model_wts = copy.deepcopy(model.state_dict())
                state = {
                  'state_dict': model.state_dict(),#字典里key就是各层的名字，值就是训练好的权重
                  'best_acc': best_acc,
                  'optimizer' : optimizer.state_dict(),
                }
                torch.save(state, filename)
            if phase == 'valid':
                val_acc_history.append(epoch_acc)
                valid_losses.append(epoch_loss)
                #scheduler.step(epoch_loss)#学习率衰减
            if phase == 'train':
                train_acc_history.append(epoch_acc)
                train_losses.append(epoch_loss)
        
        print('Optimizer learning rate : {:.7f}'.format(optimizer.param_groups[0]['lr']))
        LRs.append(optimizer.param_groups[0]['lr'])
        print()
        scheduler.step()#学习率衰减

    time_elapsed = time.time() - since
    print('Training complete in {:.0f}m {:.0f}s'.format(time_elapsed // 60, time_elapsed % 60))
    print('Best val Acc: {:4f}'.format(best_acc))

    # 训练完后用最好的一次当做模型最终的结果,等着一会测试
    model.load_state_dict(best_model_wts)
    return model, val_acc_history, train_acc_history, valid_losses, train_losses, LRs 

第六步：补充，使用全部的参数更新策略，将所有层的梯度更新设为true即可

for param in model_ft.parameters():
    param.requires_grad = True

# 再继续训练所有的参数，学习率调小一点
optimizer = optim.Adam(model_ft.parameters(), lr=1e-3)
scheduler = optim.lr_scheduler.StepLR(optimizer_ft, step_size=7, gamma=0.1)

# 损失函数
criterion = nn.CrossEntropyLoss()

第七步：用之前保存的模型进行训练，就是所谓的迁移学习，将他的权重参数全部导入进来

model_ft, input_size = initialize_model(model_name, 102, feature_extract, use_pretrained=True)

# GPU模式
model_ft = model_ft.to(device)

# 保存文件的名字
filename='best.pt'

# 加载模型
checkpoint = torch.load(filename)
best_acc = checkpoint['best_acc']
model_ft.load_state_dict(checkpoint['state_dict'])

第八步：进行测试

# 得到一个batch的测试数据
dataiter = iter(dataloaders['valid'])
images, labels = dataiter.next()

model_ft.eval()

if train_on_gpu:
    output = model_ft(images.cuda())
else:
    output = model_ft(images)
_, preds_tensor = torch.max(output, 1)

preds = np.squeeze(preds_tensor.numpy()) if not train_on_gpu else np.squeeze(preds_tensor.cpu().numpy())

def im_convert(tensor):
    """ 展示数据"""
    
    image = tensor.to("cpu").clone().detach()
    image = image.numpy().squeeze()
    image = image.transpose(1,2,0)
    image = image * np.array((0.229, 0.224, 0.225)) + np.array((0.485, 0.456, 0.406))
    image = image.clip(0, 1)

    return image
fig=plt.figure(figsize=(20, 20))
columns =4
rows = 2

for idx in range (columns*rows):
    ax = fig.add_subplot(rows, columns, idx+1, xticks=[], yticks=[])
    plt.imshow(im_convert(images[idx]))
    ax.set_title("{} ({})".format(cat_to_name[str(preds[idx])], cat_to_name[str(labels[idx].item())]),
                 color=("green" if cat_to_name[str(preds[idx])]==cat_to_name[str(labels[idx].item())] else "red"))
plt.show()

流程总结：