torch 一文遷移學習　蜜蜂螞蟻　BeeAndAnt

本文實際上譯自 Transfer Learning Tutorial,嚴重參考了斯坦福有關Transfer Learning的介紹

這篇博客將詳細介紹:

1. 介紹使用預訓練模型的兩種方法（fine-tuning及fixed feature extractor）

2. pytorch官方給出的對應的樣例程序

１．介紹使用預訓練模型的兩種方法（fine-tuning及fixed feature extractor）

簡單來說，fixed feature extractor只訓練最後的隱藏層與輸出層（可以去掉原模型的輸出層，加入自己的輸出層）之間的參數，或者乾脆不進行訓練，把最後一層隱藏層作爲特徵向量輸出，從而成爲一個特徵提取器。而fine-tuning可以訓練之前隱藏層的參數。

使用情況：上述斯坦福的文章中討論了四種情況下（數據集大小，數據集與預訓練模型使用的數據集之間的差別大小），預訓練模型應該怎麼使用。fixed feature extractor適合於數據集較小，數據集與預訓練模型使用的數據集之間的差別不大的情況，而fine-tuning適合於數據集較大，數據集與預訓練模型使用的數據集之間的差別不大的情況。

詳細內容，推薦閱讀斯坦福的文章。

２．pytorch官方給出的對應的樣例程序

1. 首先，導入具體的庫

２．構造數據集

首先下載數據集，點擊這裏，解壓後把裏面的train及valid文件夾放在當前工作目錄。包括對數據進行數據增強（Data augmentation），包括進行隨機固定尺寸裁剪（預訓練模型對深入圖片的大小有要求），進行隨機旋轉鏡像，以及規範化，這些操作多有明確的好處（可理解成固定的優秀的套路，祖傳操作，，，），具體介紹請參考cs231n課程。

# Data augmentation and normalization for training
# Just normalization for validation
data_transforms = {
    'train': transforms.Compose([
        transforms.RandomResizedCrop(224),
        transforms.RandomHorizontalFlip(),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ]),
    'val': transforms.Compose([
        transforms.Resize(256),
        transforms.CenterCrop(224),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ]),
}

data_dir = ''
image_datasets = {x: datasets.ImageFolder(os.path.join(data_dir, x),
                                          data_transforms[x])
                  for x in ['train', 'val']}
dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=4,
                                             shuffle=True, num_workers=4)
              for x in ['train', 'val']}
dataset_sizes = {x: len(image_datasets[x]) for x in ['train', 'val']}
class_names = image_datasets['train'].classes

３．先看一波圖片樣子

def imshow(inp, title=None):
    """Imshow for Tensor."""
    inp = inp.numpy().transpose((1, 2, 0))
    mean = np.array([0.485, 0.456, 0.406])
    std = np.array([0.229, 0.224, 0.225])
    inp = std * inp + mean
    inp = np.clip(inp, 0, 1)
    plt.imshow(inp)
    if title is not None:
        plt.title(title)
    plt.pause(0.001 )  # 如果感覺停留時間短可以調整0.001單位秒


# Get a batch of training data
inputs, classes = next(iter(dataloaders['train']))

# Make a grid from batch
out = torchvision.utils.make_grid(inputs)
imshow(out, title=[class_names[x] for x in classes])
plt.show()

　　　　　　　　　　　　　 　

４．模型訓練及可視化模板

def train_model(model, criterion, optimizer, scheduler, num_epochs=25):
    since = time.time()

    best_model_wts = copy.deepcopy(model.state_dict())
    best_acc = 0.0

    for epoch in range(num_epochs):
        print('Epoch {}/{}'.format(epoch, num_epochs - 1))
        print('-' * 10)

        # Each epoch has a training and validation phase
        for phase in ['train', 'val']:
            if phase == 'train':
                scheduler.step()
                model.train()  # Set model to training mode
            else:
                model.eval()   # Set model to evaluate mode

            running_loss = 0.0
            running_corrects = 0

            # Iterate over data.
            for inputs, labels in dataloaders[phase]:
                inputs = inputs.to(device)
                labels = labels.to(device)

                # zero the parameter gradients
                optimizer.zero_grad()

                # forward
                # track history if only in train
                with torch.set_grad_enabled(phase == 'train'):
                    outputs = model(inputs)
                    _, preds = torch.max(outputs, 1)
                    loss = criterion(outputs, labels)

                    # backward + optimize only if in training phase
                    if phase == 'train':
                        loss.backward()
                        optimizer.step()

                # statistics
                running_loss += loss.item() * inputs.size(0)
                running_corrects += torch.sum(preds == labels.data)

            epoch_loss = running_loss / dataset_sizes[phase]
            epoch_acc = running_corrects.double() / dataset_sizes[phase]

            print('{} Loss: {:.4f} Acc: {:.4f}'.format(
                phase, epoch_loss, epoch_acc))

            # deep copy the model
            if phase == 'val' and epoch_acc > best_acc:
                best_acc = epoch_acc
                best_model_wts = copy.deepcopy(model.state_dict())

        print()

    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))

    # load best model weights
    model.load_state_dict(best_model_wts)
    return model
def visualize_model(model, num_images=6):
    was_training = model.training
    model.eval()
    images_so_far = 0
    fig = plt.figure()

    with torch.no_grad():
        for i, (inputs, labels) in enumerate(dataloaders['val']):
            inputs = inputs.to(device)
            labels = labels.to(device)

            outputs = model(inputs)
            _, preds = torch.max(outputs, 1)

            for j in range(inputs.size()[0]):
                images_so_far += 1
                ax = plt.subplot(num_images//2, 2, images_so_far)
                ax.axis('off')
                ax.set_title('predicted: {}'.format(class_names[preds[j]]))
                imshow(inputs.cpu().data[j])

                if images_so_far == num_images:
                    model.train(mode=was_training)
                    return
        model.train(mode=was_training)

５．如果你有ＧＰＵ可以使用，那將極大的提高你的訓練速度，設置使用ＧＰＵ的方法如下:

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")

６．１構建模型 fine-tuning，在最後隱藏層接輸出層處斷開，然後姐兩類輸出。

注意，一般模型最後的一層爲全連接層（fully connected layer），簡稱　fc　。model_ft.fc.in_features是該層輸入的神經元（特徵）數。這裏並沒有凍結前面層的參數。

model_ft = models.resnet18(pretrained=True)
num_ftrs = model_ft.fc.in_features
model_ft.fc = nn.Linear(num_ftrs, 2)

model_ft = model_ft.to(device)

７．１指定評價函數，優化器，及訓練計劃開始訓練 

criterion = nn.CrossEntropyLoss()

# Observe that all parameters are being optimized
optimizer_ft = optim.SGD(model_ft.parameters(), lr=0.001, momentum=0.9)

# Decay LR by a factor of 0.1 every 7 epochs
exp_lr_scheduler = lr_scheduler.StepLR(optimizer_ft, step_size=7, gamma=0.1)
model_ft = train_model(model_ft, criterion, optimizer_ft, exp_lr_scheduler,num_epochs=25)
# visualize_model(model_ft)

６．２構建模型 fine-tuning，在最後隱藏層接輸出層處斷開，然後姐兩類輸出。

注意，一般模型最後的一層爲全連接層（fully connected layer），簡稱　fc　。model_ft.fc.in_features是該層輸入的神經元（特徵）數。並通過  param.requires_grad = False　凍結前面層的參數（即訓練時，不會更新這些參數）。

model_conv = torchvision.models.resnet18(pretrained=True)
for param in model_conv.parameters():
    param.requires_grad = False

# Parameters of newly constructed modules have requires_grad=True by default
num_ftrs = model_conv.fc.in_features
model_conv.fc = nn.Linear(num_ftrs, 2)

model_conv = model_conv.to(device)

７．１指定評價函數，優化器，及訓練計劃開始訓練  

criterion = nn.CrossEntropyLoss()

# Observe that only parameters of final layer are being optimized as
# opposed to before.
optimizer_conv = optim.SGD(model_conv.fc.parameters(), lr=0.001, momentum=0.9)

# Decay LR by a factor of 0.1 every 7 epochs
exp_lr_scheduler = lr_scheduler.StepLR(optimizer_conv, step_size=7, gamma=0.1)
model_conv = train_model(model_conv, criterion, optimizer_conv,
                         exp_lr_scheduler, num_epochs=25)