Kevin's Blog
0%

PyTorch入门-2

发表于

PyTorch入门-2

Datesets & Dataloaders

位置： torch.utils.data.DataLoader and torch.utils.data.Dataset

Dataset存储样本及其对应的标签，DataLoader在数据集周围包装了一个可迭代对象，以方便访问样本

加载数据

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
import torch
from torch.utils.data import Dataset
from torchvision import datasets
from torchvision.transforms import ToTensor
import matplotlib.pyplot as plt


training_data = datasets.FashionMNIST(
    root="data",	#存储数据的路径
    train=True,		#辨别是训练集还是测试集
    download=True,	#如果不在root路径下，则需要下载
    transform=ToTensor()	#transform and target_transform specify the feature and label transformations
)

test_data = datasets.FashionMNIST(
    root="data",
    train=False,
    download=True,
    transform=ToTensor()
)

迭代、可视化数据集

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
labels_map = {
    0: "T-Shirt",
    1: "Trouser",
    2: "Pullover",
    3: "Dress",
    4: "Coat",
    5: "Sandal",
    6: "Shirt",
    7: "Sneaker",
    8: "Bag",
    9: "Ankle Boot",
}
figure = plt.figure(figsize=(8, 8))
cols, rows = 3, 3
for i in range(1, cols * rows + 1):
    sample_idx = torch.randint(len(training_data), size=(1,)).item()
    img, label = training_data[sample_idx]
    figure.add_subplot(rows, cols, i)
    plt.title(labels_map[label])
    plt.axis("off")
    plt.imshow(img.squeeze(), cmap="gray")
plt.show()

创建自定义数据集

自定义数据集类必须包含三个函数：__init__, __len__, and __getitem__

FashionMNIST图像存储在目录img_dir中，它们的标签分别存储在CSV文件annotations_file中。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
import os
import pandas as pd
from torchvision.io import read_image

class CustomImageDataset(Dataset):
    def __init__(self, annotations_file, img_dir, transform=None, target_transform=None):
        #self.img_labels = pd.read_csv(annotations_file)
        self.img_labels = pd.read_csv(annotations_file, names=['file_name', 'label'])
        self.img_dir = img_dir
        self.transform = transform
        self.target_transform = target_transform
# labels.csv格式：
# tshirt1.jpg, 0
# tshirt2.jpg, 0
# ......
# ankleboot999.jpg, 9

    def __len__(self):
        return len(self.img_labels)

    def __getitem__(self, idx):#从给定索引idx处的数据集加载并返回一个示例
        img_path = os.path.join(self.img_dir, self.img_labels.iloc[idx, 0])
        image = read_image(img_path)
        label = self.img_labels.iloc[idx, 1]
        if self.transform:
            image = self.transform(image)
        if self.target_transform:
            label = self.target_transform(label)
        return image, label

使用DataLoader为训练做准备，遍历DataLoader

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
from torch.utils.data import DataLoader

train_dataloader = DataLoader(training_data, batch_size=64, shuffle=True) #shuffle即将数据打散
test_dataloader = DataLoader(test_data, batch_size=64, shuffle=True)

# Display image and label.
train_features, train_labels = next(iter(train_dataloader))
print(f"Feature batch shape: {train_features.size()}")
print(f"Labels batch shape: {train_labels.size()}")
img = train_features[0].squeeze()
label = train_labels[0]
plt.imshow(img, cmap="gray")
plt.show()
print(f"Label: {label}")

Feature batch shape: torch.Size([64, 1, 28, 28])
Labels batch shape: torch.Size([64])
Label: 7

transforms

数据并不总是以训练机器学习算法所需的最终处理形式出现。我们使用转换来对数据执行一些操作，并使其适合于训练

transform修改特征，target_transform修改标签

1
2
3
4
5
6
7
8
9
10
11
import torch
from torchvision import datasets
from torchvision.transforms import ToTensor, Lambda

ds = datasets.FashionMNIST(
    root="data",
    train=True,
    download=True,
    transform=ToTensor(), 	#将PIL图像或Numpy数组转化为FloatTensor，且范围在[0,1]之间
    target_transform=Lambda(lambda y: torch.zeros(10, dtype=torch.float).scatter_(0, torch.tensor(y), value=1))	#将整数转化为one-hot编码，将位置处赋值为1
)