一、遷移學習定義
遷移學習（Transfer Learning）：在一個任務上訓練得到的模型包含的知識可以部分或全部地轉移到另一個任務上。允許模型將從一個任務中學到的知識應用到另一個相關的任務中。適用于數據稀缺的情況，可減少對大量標記數據的需求。
遷移學習是一種機器學習方法，具體是指將已經在某一領域（或任務）學習到的知識或模型，應用到另一個不同但相關的領域（或任務）中，以提高在該新任務上的學習效率和效果。這種知識或模型的遷移可以包括網絡參數、特征表示、數據間的關系等多種形式的知識。
遷移學習通常可以分為以下幾種類型：
- 基于模型的遷移學習：直接使用源任務的預訓練模型作為目標任務的起點，進行微調或重新訓練。
- 基于特征的遷移學習：從源任務中提取特征表示，然后在這些特征上訓練目標任務的模型。
- 基于關系的遷移學習：從源任務中學習數據間的關系，然后將這種關系應用到目標任務中。
二、遷移學習實現流程
遷移學習的實現流程通常包括以下幾個步驟：
- 選擇源任務：選擇一個具有豐富數據的相關預測建模問題作為源任務。
- 開發源模型：為源任務開發一個精巧的模型，并確保其性能優于普通模型。
- 重用模型：將源任務的模型作為目標任務的學習起點（固定特征進行訓練），這可能涉及全部或部分使用源模型。
- 調整模型：在目標數據集上對模型進行微調，以使其適應目標任務。
三、Resnet50遷移學習實踐：狗、狼分類
3.1下載數據集
狼狗數據集提取自ImageNet分類數據集
from download import downloaddataset_url = "https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/datasets/intermediate/Canidae_data.zip"download(dataset_url, "./datasets-Canidae", kind="zip", replace=True)
輸出數據集目錄結構：
使用mindspore.dataset.ImageFolderDataset接口來加載數據集，并進行相關圖像增強操作
import mindspore as ms
import mindspore.dataset as ds
import mindspore.dataset.vision as vision# 數據集目錄路徑
data_path_train = "./datasets-Canidae/data/Canidae/train/"
data_path_val = "./datasets-Canidae/data/Canidae/val/"# 創建訓練數據集def create_dataset_canidae(dataset_path, usage):"""數據加載"""data_set = ds.ImageFolderDataset(dataset_path,num_parallel_workers=workers,shuffle=True,)# 數據增強操作mean = [0.485 * 255, 0.456 * 255, 0.406 * 255]std = [0.229 * 255, 0.224 * 255, 0.225 * 255]scale = 32if usage == "train":# Define map operations for training datasettrans = [vision.RandomCropDecodeResize(size=image_size, scale=(0.08, 1.0), ratio=(0.75, 1.333)),vision.RandomHorizontalFlip(prob=0.5),vision.Normalize(mean=mean, std=std),vision.HWC2CHW()]else:# Define map operations for inference datasettrans = [vision.Decode(),vision.Resize(image_size + scale),vision.CenterCrop(image_size),vision.Normalize(mean=mean, std=std),vision.HWC2CHW()]# 數據映射操作data_set = data_set.map(operations=trans,input_columns='image',num_parallel_workers=workers)# 批量操作data_set = data_set.batch(batch_size)return data_setdataset_train = create_dataset_canidae(data_path_train, "train")
step_size_train = dataset_train.get_dataset_size()dataset_val = create_dataset_canidae(data_path_val, "val")
step_size_val = dataset_val.get_dataset_size()# 數據集可視化
data = next(dataset_train.create_dict_iterator())
images = data["image"]
labels = data["label"]print("Tensor of image", images.shape)
print("Labels:", labels)import matplotlib.pyplot as plt
import numpy as np
# class_name對應label，按文件夾字符串從小到大的順序標記label
class_name = {0: "dogs", 1: "wolves"}plt.figure(figsize=(5, 5))
for i in range(4):# 獲取圖像及其對應的labeldata_image = images[i].asnumpy()data_label = labels[i]# 處理圖像供展示使用data_image = np.transpose(data_image, (1, 2, 0))mean = np.array([0.485, 0.456, 0.406])std = np.array([0.229, 0.224, 0.225])data_image = std * data_image + meandata_image = np.clip(data_image, 0, 1)# 顯示圖像plt.subplot(2, 2, i+1)plt.imshow(data_image)plt.title(class_name[int(labels[i].asnumpy())])plt.axis("off")plt.show()
3.2 構建神經網絡模型
使用ResNet50模型進行訓練。搭建好模型框架后，通過將pretrained參數設置為True來下載ResNet50的預訓練模型并將權重參數加載到網絡中。
下載ResNet50的預訓練模型
1.構建模型
from typing import Type, Union, List, Optional
from mindspore import nn, train
from mindspore.common.initializer import Normalweight_init = Normal(mean=0, sigma=0.02)
gamma_init = Normal(mean=1, sigma=0.02)class ResidualBlockBase(nn.Cell):expansion: int = 1 # 最后一個卷積核數量與第一個卷積核數量相等def __init__(self, in_channel: int, out_channel: int,stride: int = 1, norm: Optional[nn.Cell] = None,down_sample: Optional[nn.Cell] = None) -> None:super(ResidualBlockBase, self).__init__()if not norm:self.norm = nn.BatchNorm2d(out_channel)else:self.norm = normself.conv1 = nn.Conv2d(in_channel, out_channel,kernel_size=3, stride=stride,weight_init=weight_init)self.conv2 = nn.Conv2d(in_channel, out_channel,kernel_size=3, weight_init=weight_init)self.relu = nn.ReLU()self.down_sample = down_sampledef construct(self, x):"""ResidualBlockBase construct."""identity = x # shortcuts分支out = self.conv1(x) # 主分支第一層：3*3卷積層out = self.norm(out)out = self.relu(out)out = self.conv2(out) # 主分支第二層：3*3卷積層out = self.norm(out)if self.down_sample is not None:identity = self.down_sample(x)out += identity # 輸出為主分支與shortcuts之和out = self.relu(out)return out
殘差模塊：
class ResidualBlock(nn.Cell):expansion = 4 # 最后一個卷積核的數量是第一個卷積核數量的4倍def __init__(self, in_channel: int, out_channel: int,stride: int = 1, down_sample: Optional[nn.Cell] = None) -> None:super(ResidualBlock, self).__init__()self.conv1 = nn.Conv2d(in_channel, out_channel,kernel_size=1, weight_init=weight_init)self.norm1 = nn.BatchNorm2d(out_channel)self.conv2 = nn.Conv2d(out_channel, out_channel,kernel_size=3, stride=stride,weight_init=weight_init)self.norm2 = nn.BatchNorm2d(out_channel)self.conv3 = nn.Conv2d(out_channel, out_channel * self.expansion,kernel_size=1, weight_init=weight_init)self.norm3 = nn.BatchNorm2d(out_channel * self.expansion)self.relu = nn.ReLU()self.down_sample = down_sampledef construct(self, x):identity = x # shortscuts分支out = self.conv1(x) # 主分支第一層：1*1卷積層out = self.norm1(out)out = self.relu(out)out = self.conv2(out) # 主分支第二層：3*3卷積層out = self.norm2(out)out = self.relu(out)out = self.conv3(out) # 主分支第三層：1*1卷積層out = self.norm3(out)if self.down_sample is not None:identity = self.down_sample(x)out += identity # 輸出為主分支與shortcuts之和out = self.relu(out)return out
網絡層
def make_layer(last_out_channel, block: Type[Union[ResidualBlockBase, ResidualBlock]],channel: int, block_nums: int, stride: int = 1):down_sample = None # shortcuts分支if stride != 1 or last_out_channel != channel * block.expansion:down_sample = nn.SequentialCell([nn.Conv2d(last_out_channel, channel * block.expansion,kernel_size=1, stride=stride, weight_init=weight_init),nn.BatchNorm2d(channel * block.expansion, gamma_init=gamma_init)])layers = []layers.append(block(last_out_channel, channel, stride=stride, down_sample=down_sample))in_channel = channel * block.expansion# 堆疊殘差網絡for _ in range(1, block_nums):layers.append(block(in_channel, channel))return nn.SequentialCell(layers)組合
from mindspore import load_checkpoint, load_param_into_netclass ResNet(nn.Cell):def __init__(self, block: Type[Union[ResidualBlockBase, ResidualBlock]],layer_nums: List[int], num_classes: int, input_channel: int) -> None:super(ResNet, self).__init__()self.relu = nn.ReLU()# 第一個卷積層，輸入channel為3（彩色圖像），輸出channel為64self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, weight_init=weight_init)self.norm = nn.BatchNorm2d(64)# 最大池化層，縮小圖片的尺寸self.max_pool = nn.MaxPool2d(kernel_size=3, stride=2, pad_mode='same')# 各個殘差網絡結構塊定義，self.layer1 = make_layer(64, block, 64, layer_nums[0])self.layer2 = make_layer(64 * block.expansion, block, 128, layer_nums[1], stride=2)self.layer3 = make_layer(128 * block.expansion, block, 256, layer_nums[2], stride=2)self.layer4 = make_layer(256 * block.expansion, block, 512, layer_nums[3], stride=2)# 平均池化層self.avg_pool = nn.AvgPool2d()# flattern層self.flatten = nn.Flatten()# 全連接層self.fc = nn.Dense(in_channels=input_channel, out_channels=num_classes)def construct(self, x):x = self.conv1(x)x = self.norm(x)x = self.relu(x)x = self.max_pool(x)x = self.layer1(x)x = self.layer2(x)x = self.layer3(x)x = self.layer4(x)x = self.avg_pool(x)x = self.flatten(x)x = self.fc(x)return xdef _resnet(model_url: str, block: Type[Union[ResidualBlockBase, ResidualBlock]],layers: List[int], num_classes: int, pretrained: bool, pretrianed_ckpt: str,input_channel: int):model = ResNet(block, layers, num_classes, input_channel)if pretrained:# 加載預訓練模型download(url=model_url, path=pretrianed_ckpt, replace=True)param_dict = load_checkpoint(pretrianed_ckpt)load_param_into_net(model, param_dict)return modeldef resnet50(num_classes: int = 1000, pretrained: bool = False):"ResNet50模型"resnet50_url = "https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/models/application/resnet50_224_new.ckpt"resnet50_ckpt = "./LoadPretrainedModel/resnet50_224_new.ckpt"return _resnet(resnet50_url, ResidualBlock, [3, 4, 6, 3], num_classes,pretrained, resnet50_ckpt, 2048)
3.3 固定特征進行訓練
使用固定特征進行訓練的時候，需要凍結除最后一層之外的所有網絡層。通過設置requires_grad == False凍結參數，以便不在反向傳播中計算梯度。
import mindspore as ms
import matplotlib.pyplot as plt
import os
import timenet_work = resnet50(pretrained=True)# 全連接層輸入層的大小
in_channels = net_work.fc.in_channels
# 輸出通道數大小為狼狗分類數2
head = nn.Dense(in_channels, 2)
# 重置全連接層
net_work.fc = head# 平均池化層kernel size為7
avg_pool = nn.AvgPool2d(kernel_size=7)
# 重置平均池化層
net_work.avg_pool = avg_pool# 凍結除最后一層外的所有參數
for param in net_work.get_parameters():if param.name not in ["fc.weight", "fc.bias"]:param.requires_grad = False# 定義優化器和損失函數
opt = nn.Momentum(params=net_work.trainable_params(), learning_rate=lr, momentum=0.5)
loss_fn = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')def forward_fn(inputs, targets):logits = net_work(inputs)loss = loss_fn(logits, targets)return lossgrad_fn = ms.value_and_grad(forward_fn, None, opt.parameters)def train_step(inputs, targets):loss, grads = grad_fn(inputs, targets)opt(grads)return loss# 實例化模型
model1 = train.Model(net_work, loss_fn, opt, metrics={"Accuracy": train.Accuracy()})
訓練和評估
import mindspore as ms
import matplotlib.pyplot as plt
import os
import time
dataset_train = create_dataset_canidae(data_path_train, "train")
step_size_train = dataset_train.get_dataset_size()dataset_val = create_dataset_canidae(data_path_val, "val")
step_size_val = dataset_val.get_dataset_size()num_epochs = 5# 創建迭代器
data_loader_train = dataset_train.create_tuple_iterator(num_epochs=num_epochs)
data_loader_val = dataset_val.create_tuple_iterator(num_epochs=num_epochs)
best_ckpt_dir = "./BestCheckpoint"
可視化模型預測
使用固定特征得到的best.ckpt文件對對驗證集的狼和狗圖像數據進行預測。若預測字體為藍色即為預測正確，若預測字體為紅色則預測錯誤。
import mindspore as ms
import matplotlib.pyplot as plt
import os
import time
# 開始循環訓練
print("Start Training Loop ...")best_acc = 0for epoch in range(num_epochs):losses = []net_work.set_train()epoch_start = time.time()# 為每輪訓練讀入數據for i, (images, labels) in enumerate(data_loader_train):labels = labels.astype(ms.int32)loss = train_step(images, labels)losses.append(loss)# 每個epoch結束后，驗證準確率acc = model1.eval(dataset_val)['Accuracy']epoch_end = time.time()epoch_seconds = (epoch_end - epoch_start) * 1000step_seconds = epoch_seconds/step_size_trainprint("-" * 20)print("Epoch: [%3d/%3d], Average Train Loss: [%5.3f], Accuracy: [%5.3f]" % (epoch+1, num_epochs, sum(losses)/len(losses), acc))print("epoch time: %5.3f ms, per step time: %5.3f ms" % (epoch_seconds, step_seconds))if acc > best_acc:best_acc = accif not os.path.exists(best_ckpt_dir):os.mkdir(best_ckpt_dir)ms.save_checkpoint(net_work, best_ckpt_path)print("=" * 80)
print(f"End of validation the best Accuracy is: {best_acc: 5.3f}, "f"save the best ckpt file in {best_ckpt_path}", flush=True)
import matplotlib.pyplot as plt
import mindspore as msdef visualize_model(best_ckpt_path, val_ds):net = resnet50()# 全連接層輸入層的大小in_channels = net.fc.in_channels# 輸出通道數大小為狼狗分類數2head = nn.Dense(in_channels, 2)# 重置全連接層net.fc = head# 平均池化層kernel size為7avg_pool = nn.AvgPool2d(kernel_size=7)# 重置平均池化層net.avg_pool = avg_pool# 加載模型參數param_dict = ms.load_checkpoint(best_ckpt_path)ms.load_param_into_net(net, param_dict)model = train.Model(net)# 加載驗證集的數據進行驗證data = next(val_ds.create_dict_iterator())images = data["image"].asnumpy()labels = data["label"].asnumpy()class_name = {0: "dogs", 1: "wolves"}# 預測圖像類別output = model.predict(ms.Tensor(data['image']))pred = np.argmax(output.asnumpy(), axis=1)# 顯示圖像及圖像的預測值plt.figure(figsize=(5, 5))for i in range(4):plt.subplot(2, 2, i + 1)# 若預測正確，顯示為藍色；若預測錯誤，顯示為紅色color = 'blue' if pred[i] == labels[i] else 'red'plt.title('predict:{}'.format(class_name[pred[i]]), color=color)picture_show = np.transpose(images[i], (1, 2, 0))mean = np.array([0.485, 0.456, 0.406])std = np.array([0.229, 0.224, 0.225])picture_show = std * picture_show + meanpicture_show = np.clip(picture_show, 0, 1)plt.imshow(picture_show)plt.axis('off')plt.show()visualize_model(best_ckpt_path, dataset_val)
】)
附錄B(gpt-4o翻譯版))
