- torchvision .datasets 模块中提供很多内置数据集;
- 以及很多工具类用于构建用户自己的数据集;
内置数据集
- 所有内置的数据集都是 torch.utils.data.Dataset的子类;
- 也就是他们都具有已经实现的__getitem__和__len__方法;
- 内置数据集都能够被送到torch.utils.data.DataLoader;
- 以并行的方式使用torch.multiprocessing加载多种样本;
- 代码实例:
import torch.utils.data
import torchvision.datasets
imagenet_data=torchvision.datasets.ImageNet('path/to/imagenet_root/')
data_loader=torch.utils.data.DataLoader(imagenet_data,
batch_size=4,
shuffle=True,
num_workers=args.nThreads)
- 所有的数据集具有类似的 API;
- 所有的 API 都具有两个共同的参数:transform 和 target_transform,独立的转换输入和目标;
- 使用 pytorch 提供的基础类用户可以创建自己的数据集 ;
图像分类数据集
Caltech101(root[, target_type, transform, ...]) | Caltech 101 Dataset. |
Caltech256(root[, transform, ...]) | Caltech 256 Dataset. |
CelebA(root[, split, target_type, ...]) | |
CIFAR10(root[, train, transform, ...]) | CIFAR10 Dataset. |
CIFAR100(root[, train, transform, ...]) | CIFAR100 Dataset. |
Country211(root[, split, transform, ...]) | The Country211 Data Set from OpenAI. |
DTD(root[, split, partition, transform, ...]) | |
EMNIST(root, split, **kwargs) | EMNIST Dataset. |
EuroSAT(root[, transform, target_transform, ...]) | RGB version of the EuroSAT Dataset. |
FakeData([size, image_size, num_classes, ...]) | A fake dataset that returns randomly generated images and returns them as PIL images |
FashionMNIST(root[, train, transform, ...]) | Fashion-MNIST Dataset. |
FER2013(root[, split, transform, ...]) | FER2013 Dataset. |
FGVCAircraft(root[, split, ...]) | FGVC Aircraft Dataset. |
Flickr8k(root, ann_file[, transform, ...]) | Flickr8k Entities Dataset. |
Flickr30k(root, ann_file[, transform, ...]) | Flickr30k Entities Dataset. |
Flowers102(root[, split, transform, ...]) | Oxford 102 Flower Dataset. |
Food101(root[, split, transform, ...]) | |
GTSRB(root[, split, transform, ...]) | |
INaturalist(root[, version, target_type, ...]) | iNaturalist Dataset. |
ImageNet(root[, split]) | ImageNet 2012 Classification Dataset. |
Imagenette(root[, split, size, download, ...]) | Imagenette image classification dataset. |
KMNIST(root[, train, transform, ...]) | Kuzushiji-MNIST Dataset. |
LFWPeople(root[, split, image_set, ...]) | LFW Dataset. |
LSUN(root[, classes, transform, ...]) | LSUN dataset. |
MNIST(root[, train, transform, ...]) | MNIST Dataset. |
Omniglot(root[, background, transform, ...]) | Omniglot Dataset. |
OxfordIIITPet(root[, split, target_types, ...]) | |
Places365(root, split, small, download, ...) | Places365 classification dataset. |
PCAM(root[, split, transform, ...]) | |
QMNIST(root[, what, compat, train]) | QMNIST Dataset. |
RenderedSST2(root[, split, transform, ...]) | |
SEMEION(root[, transform, target_transform, ...]) | SEMEION Dataset. |
SBU(root[, transform, target_transform, ...]) | SBU Captioned Photo Dataset. |
StanfordCars(root[, split, transform, ...]) | Stanford Cars Dataset |
STL10(root[, split, folds, transform, ...]) | STL10 Dataset. |
SUN397(root[, transform, target_transform, ...]) | |
SVHN(root[, split, transform, ...]) | SVHN Dataset. |
USPS(root[, train, transform, ...]) | USPS Dataset. |
图像探测和分割数据集
CocoDetection(root, annFile[, transform, ...]) | MS Coco Detection Dataset. |
CelebA(root[, split, target_type, ...]) | |
Cityscapes(root[, split, mode, target_type, ...]) | Cityscapes Dataset. |
Kitti(root[, train, transform, ...]) | KITTI Dataset. |
OxfordIIITPet(root[, split, target_types, ...]) | |
SBDataset(root[, image_set, mode, download, ...]) | |
VOCSegmentation(root[, year, image_set, ...]) | Pascal VOC Segmentation Dataset. |
VOCDetection(root[, year, image_set, ...]) | Pascal VOC Detection Dataset. |
WIDERFace(root[, split, transform, ...]) | WIDERFace Dataset. |
光流数据集
FlyingChairs(root[, split, transforms]) | FlyingChairs Dataset for optical flow. |
FlyingThings3D(root[, split, pass_name, ...]) | FlyingThings3D dataset for optical flow. |
HD1K(root[, split, transforms]) | HD1K dataset for optical flow. |
KittiFlow(root[, split, transforms]) | KITTI dataset for optical flow (2015). |
Sintel(root[, split, pass_name, transforms]) | Sintel Dataset for optical flow. |
立体匹配数据集
CarlaStereo(root[, transforms]) | Carla simulator data linked in the CREStereo github repo. |
Kitti2012Stereo(root[, split, transforms]) | KITTI dataset from the 2012 stereo evaluation benchmark. |
Kitti2015Stereo(root[, split, transforms]) | KITTI dataset from the 2015 stereo evaluation benchmark. |
CREStereo(root[, transforms]) | Synthetic dataset used in training the CREStereo architecture. |
FallingThingsStereo(root[, variant, transforms]) | FallingThings dataset. |
SceneFlowStereo(root[, variant, pass_name, ...]) | Dataset interface for Scene Flow datasets. |
SintelStereo(root[, pass_name, transforms]) | Sintel Stereo Dataset. |
InStereo2k(root[, split, transforms]) | InStereo2k dataset. |
ETH3DStereo(root[, split, transforms]) | ETH3D Low-Res Two-View dataset. |
Middlebury2014Stereo(root[, split, ...]) | Publicly available scenes from the Middlebury dataset 2014 version <https://vision.middlebury.edu/stereo/data/scenes2014/>. |
图像配对数据集
LFWPairs(root[, split, image_set, ...]) | LFW Dataset. |
PhotoTour(root, name[, train, transform, ...]) | Multi-view Stereo Correspondence Dataset. |
图像说明数据集
CocoCaptions(root, annFile[, transform, ...]) | MS Coco Captions Dataset. |
视频分类数据集
HMDB51(root, annotation_path, frames_per_clip) | HMDB51 dataset. |
Kinetics(root, frames_per_clip[, ...]) | Generic Kinetics dataset. |
UCF101(root, annotation_path, frames_per_clip) | UCF101 dataset. |
视频预测数据集
MovingMNIST(root[, split, split_ratio, ...]) | MovingMNIST Dataset. |
用于定制数据集的基础类
DatasetFolder(root, loader[, extensions, ...]) | A generic data loader. |
ImageFolder(root, transform, ...) | A generic data loader where the images are arranged in this way by default: . |
VisionDataset([root, transforms, transform, ...]) | Base Class For making datasets which are compatible with torchvision. |
V2
wrap_dataset_for_transforms_v2(dataset[, ...]) | Wrap a |