torchvision.datasets¶

All datasets are subclasses of torch.utils.data.Dataset i.e, they have __getitem__ and __len__ methods implemented. Hence, they can all be passed to a torch.utils.data.DataLoader which can load multiple samples in parallel using torch.multiprocessing workers. For example:

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)

All the datasets have almost similar API. They all have two common arguments: transform and target_transform to transform the input and target respectively. You can also create your own datasets using the provided base classes.

Caltech¶

class torchvision.datasets.Caltech101(root: str, target_type: Union[List[str], str] = 'category', transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

Caltech 101 Dataset.

Warning

This class needs scipy to load target files from .mat format.

Parameters

root (string) – Root directory of dataset where directory caltech101 exists or will be saved to if download is set to True.
target_type (string or list, optional) – Type of target to use, category or annotation. Can also be a list to output a tuple with all specified target types. category represents the target class, and annotation is a list of points from a hand-generated outline. Defaults to category.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where the type of target specified by target_type.
Return type: tuple

class torchvision.datasets.Caltech256(root: str, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

Caltech 256 Dataset.

Parameters

root (string) – Root directory of dataset where directory caltech256 exists or will be saved to if download is set to True.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where target is index of the target class.
Return type: tuple

CelebA¶

class torchvision.datasets.CelebA(root: str, split: str = 'train', target_type: Union[List[str], str] = 'attr', transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

Large-scale CelebFaces Attributes (CelebA) Dataset Dataset.

Parameters

root (string) – Root directory where images are downloaded to.
split (string) – One of {‘train’, ‘valid’, ‘test’, ‘all’}. Accordingly dataset is selected.
target_type (string or list, optional) –
Type of target to use, attr, identity, bbox, or landmarks. Can also be a list to output a tuple with all specified target types. The targets represent:
- attr (np.array shape=(40,) dtype=int): binary (0, 1) labels for attributes
- identity (int): label for each person (data points with the same identity are the same person)
- bbox (np.array shape=(4,) dtype=int): bounding box (x, y, width, height)
- landmarks (np.array shape=(10,) dtype=int): landmark points (lefteye_x, lefteye_y, righteye_x, righteye_y, nose_x, nose_y, leftmouth_x, leftmouth_y, rightmouth_x, rightmouth_y)
Defaults to attr. If empty, None will be returned as target.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.ToTensor
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: Sample and meta data, optionally transformed by the respective transforms.
Return type: (Any)

CIFAR¶

class torchvision.datasets.CIFAR10(root: str, train: bool = True, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

CIFAR10 Dataset.

Parameters

root (string) – Root directory of dataset where directory cifar-10-batches-py exists or will be saved to if download is set to True.
train (bool, optional) – If True, creates dataset from training set, otherwise creates from test set.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where target is index of the target class.
Return type: tuple

class torchvision.datasets.CIFAR100(root: str, train: bool = True, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

CIFAR100 Dataset.

This is a subclass of the CIFAR10 Dataset.

Cityscapes¶

Note

Requires Cityscape to be downloaded.

class torchvision.datasets.Cityscapes(root: str, split: str = 'train', mode: str = 'fine', target_type: Union[List[str], str] = 'instance', transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, transforms: Optional[Callable] = None)[source]¶

Cityscapes Dataset.

Parameters

root (string) – Root directory of dataset where directory leftImg8bit and gtFine or gtCoarse are located.
split (string, optional) – The image split to use, train, test or val if mode=”fine” otherwise train, train_extra or val
mode (string, optional) – The quality mode to use, fine or coarse
target_type (string or list, optional) – Type of target to use, instance, semantic, polygon or color. Can also be a list to output a tuple with all specified target types.
transform (callable, optional) – A function/transform that takes in a PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
transforms (callable, optional) – A function/transform that takes input sample and its target as entry and returns a transformed version.

Examples

Get semantic segmentation target

dataset = Cityscapes('./data/cityscapes', split='train', mode='fine',
                     target_type='semantic')

img, smnt = dataset[0]

Get multiple targets

dataset = Cityscapes('./data/cityscapes', split='train', mode='fine',
                     target_type=['instance', 'color', 'polygon'])

img, (inst, col, poly) = dataset[0]

Validate on the “coarse” set

dataset = Cityscapes('./data/cityscapes', split='val', mode='coarse',
                     target_type='semantic')

img, smnt = dataset[0]

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where target is a tuple of all target types if target_type is a list with more than one item. Otherwise target is a json object if target_type=”polygon”, else the image segmentation.
Return type: tuple

COCO¶

Note

These require the COCO API to be installed

Captions¶

class torchvision.datasets.CocoCaptions(root: str, annFile: str, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, transforms: Optional[Callable] = None)[source]¶

MS Coco Captions Dataset.

Parameters

root (string) – Root directory where images are downloaded to.
annFile (string) – Path to json annotation file.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.ToTensor
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
transforms (callable, optional) – A function/transform that takes input sample and its target as entry and returns a transformed version.

Example

import torchvision.datasets as dset
import torchvision.transforms as transforms
cap = dset.CocoCaptions(root = 'dir where images are',
                        annFile = 'json annotation file',
                        transform=transforms.ToTensor())

print('Number of samples: ', len(cap))
img, target = cap[3] # load 4th sample

print("Image Size: ", img.size())
print(target)

Output:

Number of samples: 82783
Image Size: (3L, 427L, 640L)
[u'A plane emitting smoke stream flying over a mountain.',
u'A plane darts across a bright blue sky behind a mountain covered in snow',
u'A plane leaves a contrail above the snowy mountain top.',
u'A mountain that has a plane flying overheard in the distance.',
u'A mountain view with a plume of smoke in the background']

__getitem__(index: int) → Tuple[Any, Any]¶

Parameters: index (int) – Index
Returns: Sample and meta data, optionally transformed by the respective transforms.
Return type: (Any)

Detection¶

class torchvision.datasets.CocoDetection(root: str, annFile: str, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, transforms: Optional[Callable] = None)[source]¶

MS Coco Detection Dataset.

Parameters

root (string) – Root directory where images are downloaded to.
annFile (string) – Path to json annotation file.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.ToTensor
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
transforms (callable, optional) – A function/transform that takes input sample and its target as entry and returns a transformed version.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: Sample and meta data, optionally transformed by the respective transforms.
Return type: (Any)

EMNIST¶

class torchvision.datasets.EMNIST(root: str, split: str, **kwargs: Any)[source]¶

EMNIST Dataset.

Parameters

root (string) – Root directory of dataset where EMNIST/processed/training.pt and EMNIST/processed/test.pt exist.
split (string) – The dataset has 6 different splits: byclass, bymerge, balanced, letters, digits and mnist. This argument specifies which one to use.
train (bool, optional) – If True, creates dataset from training.pt, otherwise from test.pt.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.

FakeData¶

class torchvision.datasets.FakeData(size: int = 1000, image_size: Tuple[int, int, int] = (3, 224, 224), num_classes: int = 10, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, random_offset: int = 0)[source]¶

A fake dataset that returns randomly generated images and returns them as PIL images

Parameters

size (int, optional) – Size of the dataset. Default: 1000 images
image_size (tuple, optional) – Size if the returned images. Default: (3, 224, 224)
num_classes (int, optional) – Number of classes in the dataset. Default: 10
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
random_offset (int) – Offsets the index-based random seed used to generate each image. Default: 0

Fashion-MNIST¶

class torchvision.datasets.FashionMNIST(root: str, train: bool = True, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

Fashion-MNIST Dataset.

Parameters

root (string) – Root directory of dataset where FashionMNIST/processed/training.pt and FashionMNIST/processed/test.pt exist.
train (bool, optional) – If True, creates dataset from training.pt, otherwise from test.pt.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.

Flickr¶

class torchvision.datasets.Flickr8k(root: str, ann_file: str, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None)[source]¶

Flickr8k Entities Dataset.

Parameters

root (string) – Root directory where images are downloaded to.
ann_file (string) – Path to annotation file.
transform (callable, optional) – A function/transform that takes in a PIL image and returns a transformed version. E.g, transforms.ToTensor
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: Tuple (image, target). target is a list of captions for the image.
Return type: tuple

class torchvision.datasets.Flickr30k(root: str, ann_file: str, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None)[source]¶

Flickr30k Entities Dataset.

Parameters

root (string) – Root directory where images are downloaded to.
ann_file (string) – Path to annotation file.
transform (callable, optional) – A function/transform that takes in a PIL image and returns a transformed version. E.g, transforms.ToTensor
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: Tuple (image, target). target is a list of captions for the image.
Return type: tuple

HMDB51¶

class torchvision.datasets.HMDB51(root: str, annotation_path: str, frames_per_clip: int, step_between_clips: int = 1, frame_rate: Optional[int] = None, fold: int = 1, train: bool = True, transform: Optional[Callable] = None, _precomputed_metadata: Optional[Dict[str, Any]] = None, num_workers: int = 1, _video_width: int = 0, _video_height: int = 0, _video_min_dimension: int = 0, _audio_samples: int = 0)[source]¶

HMDB51 dataset.

HMDB51 is an action recognition video dataset. This dataset consider every video as a collection of video clips of fixed size, specified by frames_per_clip, where the step in frames between each clip is given by step_between_clips.

To give an example, for 2 videos with 10 and 15 frames respectively, if frames_per_clip=5 and step_between_clips=5, the dataset size will be (2 + 3) = 5, where the first two elements will come from video 1, and the next three elements from video 2. Note that we drop clips which do not have exactly frames_per_clip elements, so not all frames in a video might be present.

Internally, it uses a VideoClips object to handle clip creation.

Parameters

root (string) – Root directory of the HMDB51 Dataset.
annotation_path (str) – Path to the folder containing the split files.
frames_per_clip (int) – Number of frames in a clip.
step_between_clips (int) – Number of frames between each clip.
fold (int, optional) – Which fold to use. Should be between 1 and 3.
train (bool, optional) – If True, creates a dataset from the train split, otherwise from the test split.
transform (callable, optional) – A function/transform that takes in a TxHxWxC video and returns a transformed version.

Returns

A 3-tuple with the following entries:

video (Tensor[T, H, W, C]): The T video frames

audio(Tensor[K, L]): the audio frames, where K is the number of channels and L is the number of points

label (int): class of the video clip

Return type

tuple

__getitem__(idx: int) → Tuple[torch.Tensor, torch.Tensor, int][source]¶

Parameters: index (int) – Index
Returns: Sample and meta data, optionally transformed by the respective transforms.
Return type: (Any)

ImageNet¶

class torchvision.datasets.ImageNet(root: str, split: str = 'train', download: Optional[str] = None, **kwargs: Any)[source]¶

ImageNet 2012 Classification Dataset.

Parameters

root (string) – Root directory of the ImageNet Dataset.
split (string, optional) – The dataset split, supports train, or val.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
loader – A function to load an image given its path.

Note

This requires scipy to be installed

iNaturalist¶

class torchvision.datasets.INaturalist(root: str, version: str = '2021_train', target_type: Union[List[str], str] = 'full', transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

iNaturalist Dataset.

Parameters

root (string) – Root directory of dataset where the image files are stored. This class does not require/use annotation files.
version (string, optional) – Which version of the dataset to download/use. One of ‘2017’, ‘2018’, ‘2019’, ‘2021_train’, ‘2021_train_mini’, ‘2021_valid’. Default: 2021_train.
target_type (string or list, optional) –
Type of target to use, for 2021 versions, one of:
- full: the full category (species)
- kingdom: e.g. “Animalia”
- phylum: e.g. “Arthropoda”
- class: e.g. “Insecta”
- order: e.g. “Coleoptera”
- family: e.g. “Cleridae”
- genus: e.g. “Trichodes”
for 2017-2019 versions, one of:
- full: the full (numeric) category
- super: the super category, e.g. “Amphibians”
Can also be a list to output a tuple with all specified target types. Defaults to full.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where the type of target specified by target_type.
Return type: tuple

category_name(category_type: str, category_id: int) → str [source]¶

Parameters

category_type (str) – one of “full”, “kingdom”, “phylum”, “class”, “order”, “family”, “genus” or “super”
category_id (int) – an index (class id) from this category

Returns

the name of the category

Kinetics-400¶

class torchvision.datasets.Kinetics400(root: str, frames_per_clip: int, num_classes: Optional[Any] = None, split: Optional[Any] = None, download: Optional[Any] = None, num_download_workers: Optional[Any] = None, **kwargs: Any)[source]¶

Kinetics-400 dataset.

Kinetics-400 is an action recognition video dataset. This dataset consider every video as a collection of video clips of fixed size, specified by frames_per_clip, where the step in frames between each clip is given by step_between_clips.

To give an example, for 2 videos with 10 and 15 frames respectively, if frames_per_clip=5 and step_between_clips=5, the dataset size will be (2 + 3) = 5, where the first two elements will come from video 1, and the next three elements from video 2. Note that we drop clips which do not have exactly frames_per_clip elements, so not all frames in a video might be present.

Internally, it uses a VideoClips object to handle clip creation.

Parameters

root (string) –

Root directory of the Kinetics-400 Dataset. Should be structured as follows:

root/
├── class1
│   ├── clip1.avi
│   ├── clip2.avi
│   ├── clip3.mp4
│   └── ...
└── class2
    ├── clipx.avi
    └── ...

frames_per_clip (int) – number of frames in a clip
step_between_clips (int) – number of frames between each clip
transform (callable, optional) – A function/transform that takes in a TxHxWxC video and returns a transformed version.

Returns

A 3-tuple with the following entries:

video (Tensor[T, H, W, C]): the T video frames

audio(Tensor[K, L]): the audio frames, where K is the number of channels and L is the number of points

label (int): class of the video clip

Return type

tuple

__getitem__(idx: int) → Tuple[torch.Tensor, torch.Tensor, int]¶

Parameters: index (int) – Index
Returns: Sample and meta data, optionally transformed by the respective transforms.
Return type: (Any)

KITTI¶

class torchvision.datasets.Kitti(root: str, train: bool = True, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, transforms: Optional[Callable] = None, download: bool = False)[source]¶

KITTI Dataset.

It corresponds to the “left color images of object” dataset, for object detection.

Parameters

root (string) –

Root directory where images are downloaded to. Expects the following folder structure if download=False:

<root>
    └── Kitti
        └─ raw
            ├── training
            |   ├── image_2
            |   └── label_2
            └── testing
                └── image_2

train (bool, optional) – Use train split if true, else test split. Defaults to train.
transform (callable, optional) – A function/transform that takes in a PIL image and returns a transformed version. E.g, transforms.ToTensor
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
transforms (callable, optional) – A function/transform that takes input sample and its target as entry and returns a transformed version.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Get item at a given index.

Parameters

index (int) – Index

Returns

(image, target), where target is a list of dictionaries with the following keys:

type: str
truncated: float
occluded: int
alpha: float
bbox: float[4]
dimensions: float[3]
locations: float[3]
rotation_y: float

Return type

tuple

KMNIST¶

class torchvision.datasets.KMNIST(root: str, train: bool = True, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

Kuzushiji-MNIST Dataset.

Parameters

root (string) – Root directory of dataset where KMNIST/processed/training.pt and KMNIST/processed/test.pt exist.
train (bool, optional) – If True, creates dataset from training.pt, otherwise from test.pt.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.

LFW¶

class torchvision.datasets.LFWPeople(root: str, split: str = '10fold', image_set: str = 'funneled', transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

LFW Dataset.

Parameters

root (string) – Root directory of dataset where directory lfw-py exists or will be saved to if download is set to True.
split (string, optional) – The image split to use. Can be one of train, test, 10fold (default).
image_set (str, optional) – Type of image funneling to use, original, funneled or deepfunneled. Defaults to funneled.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomRotation
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: Tuple (image, target) where target is the identity of the person.
Return type: tuple

class torchvision.datasets.LFWPairs(root: str, split: str = '10fold', image_set: str = 'funneled', transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

LFW Dataset.

Parameters

root (string) – Root directory of dataset where directory lfw-py exists or will be saved to if download is set to True.
split (string, optional) – The image split to use. Can be one of train, test, 10fold. Defaults to 10fold.
image_set (str, optional) – Type of image funneling to use, original, funneled or deepfunneled. Defaults to funneled.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomRotation
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any, int][source]¶

Parameters: index (int) – Index
Returns: (image1, image2, target) where target is 0 for different indentities and 1 for same identities.
Return type: tuple

LSUN¶

class torchvision.datasets.LSUN(root: str, classes: Union[str, List[str]] = 'train', transform: Optional[Callable] = None, target_transform: Optional[Callable] = None)[source]¶

LSUN dataset.

You will need to install the lmdb package to use this dataset: run pip install lmdb

Parameters

root (string) – Root directory for the database files.
classes (string or list) – One of {‘train’, ‘val’, ‘test’} or a list of categories to load. e,g. [‘bedroom_train’, ‘church_outdoor_train’].
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: Tuple (image, target) where target is the index of the target category.
Return type: tuple

MNIST¶

class torchvision.datasets.MNIST(root: str, train: bool = True, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

MNIST Dataset.

Parameters

root (string) – Root directory of dataset where MNIST/processed/training.pt and MNIST/processed/test.pt exist.
train (bool, optional) – If True, creates dataset from training.pt, otherwise from test.pt.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.

Omniglot¶

class torchvision.datasets.Omniglot(root: str, background: bool = True, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

Omniglot Dataset.

Parameters

root (string) – Root directory of dataset where directory omniglot-py exists.
background (bool, optional) – If True, creates dataset from the “background” set, otherwise creates from the “evaluation” set. This terminology is defined by the authors.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset zip files from the internet and puts it in root directory. If the zip files are already downloaded, they are not downloaded again.

PhotoTour¶

class torchvision.datasets.PhotoTour(root: str, name: str, train: bool = True, transform: Optional[Callable] = None, download: bool = False)[source]¶

Multi-view Stereo Correspondence Dataset.

Note

We only provide the newer version of the dataset, since the authors state that it

is more suitable for training descriptors based on difference of Gaussian, or Harris corners, as the patches are centred on real interest point detections, rather than being projections of 3D points as is the case in the old dataset.

The original dataset is available under http://phototour.cs.washington.edu/patches/default.htm.

Parameters

root (string) – Root directory where images are.
name (string) – Name of the dataset to load.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Union[torch.Tensor, Tuple[Any, Any, torch.Tensor]][source]¶

Parameters: index (int) – Index
Returns: (data1, data2, matches)
Return type: tuple

Places365¶

class torchvision.datasets.Places365(root: str, split: str = 'train-standard', small: bool = False, download: bool = False, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, loader: Callable[[str], Any] = <function default_loader>)[source]¶

Places365 classification dataset.

Parameters

root (string) – Root directory of the Places365 dataset.
split (string, optional) – The dataset split. Can be one of train-standard (default), train-challenge, val.
small (bool, optional) – If True, uses the small images, i. e. resized to 256 x 256 pixels, instead of the high resolution ones.
download (bool, optional) – If True, downloads the dataset components and places them in root. Already downloaded archives are not downloaded again.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
loader – A function to load an image given its path.

Raises

RuntimeError – If download is False and the meta files, i. e. the devkit, are not present or corrupted.
RuntimeError – If download is True and the image archive is already extracted.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: Sample and meta data, optionally transformed by the respective transforms.
Return type: (Any)

QMNIST¶

class torchvision.datasets.QMNIST(root: str, what: Optional[str] = None, compat: bool = True, train: bool = True, **kwargs: Any)[source]¶

QMNIST Dataset.

Parameters

root (string) – Root directory of dataset whose processed subdir contains torch binary files with the datasets.
what (string,optional) – Can be ‘train’, ‘test’, ‘test10k’, ‘test50k’, or ‘nist’ for respectively the mnist compatible training set, the 60k qmnist testing set, the 10k qmnist examples that match the mnist testing set, the 50k remaining qmnist testing examples, or all the nist digits. The default is to select ‘train’ or ‘test’ according to the compatibility argument ‘train’.
compat (bool,optional) – A boolean that says whether the target for each example is class number (for compatibility with the MNIST dataloader) or a torch vector containing the full qmnist information. Default=True.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
train (bool,optional,compatibility) – When argument ‘what’ is not specified, this boolean decides whether to load the training set ot the testing set. Default: True.

SBD¶

class torchvision.datasets.SBDataset(root: str, image_set: str = 'train', mode: str = 'boundaries', download: bool = False, transforms: Optional[Callable] = None)[source]¶

Semantic Boundaries Dataset

The SBD currently contains annotations from 11355 images taken from the PASCAL VOC 2011 dataset.

Note

Please note that the train and val splits included with this dataset are different from the splits in the PASCAL VOC dataset. In particular some “train” images might be part of VOC2012 val. If you are interested in testing on VOC 2012 val, then use image_set=’train_noval’, which excludes all val images.

Warning

This class needs scipy to load target files from .mat format.

Parameters

root (string) – Root directory of the Semantic Boundaries Dataset
image_set (string, optional) – Select the image_set to use, train, val or train_noval. Image set train_noval excludes VOC 2012 val images.
mode (string, optional) – Select target type. Possible values ‘boundaries’ or ‘segmentation’. In case of ‘boundaries’, the target is an array of shape [num_classes, H, W], where num_classes=20.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.
transforms (callable, optional) – A function/transform that takes input sample and its target as entry and returns a transformed version. Input sample is PIL image and target is a numpy array if mode=’boundaries’ or PIL image if mode=’segmentation’.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: Sample and meta data, optionally transformed by the respective transforms.
Return type: (Any)

SBU¶

class torchvision.datasets.SBU(root: str, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = True)[source]¶

SBU Captioned Photo Dataset.

Parameters

root (string) – Root directory of dataset where tarball SBUCaptionedPhotoDataset.tar.gz exists.
transform (callable, optional) – A function/transform that takes in a PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If True, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where target is a caption for the photo.
Return type: tuple

SEMEION¶

class torchvision.datasets.SEMEION(root: str, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = True)[source]¶

SEMEION Dataset.

Parameters

root (string) – Root directory of dataset where directory semeion.py exists.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where target is index of the target class.
Return type: tuple

STL10¶

class torchvision.datasets.STL10(root: str, split: str = 'train', folds: Optional[int] = None, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

STL10 Dataset.

Parameters

root (string) – Root directory of dataset where directory stl10_binary exists.
split (string) – One of {‘train’, ‘test’, ‘unlabeled’, ‘train+unlabeled’}. Accordingly dataset is selected.
folds (int, optional) – One of {0-9} or None. For training, loads one of the 10 pre-defined folds of 1k samples for the standard evaluation procedure. If no value is passed, loads the 5k samples.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where target is index of the target class.
Return type: tuple

SVHN¶

class torchvision.datasets.SVHN(root: str, split: str = 'train', transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

SVHN Dataset. Note: The SVHN dataset assigns the label 10 to the digit 0. However, in this Dataset, we assign the label 0 to the digit 0 to be compatible with PyTorch loss functions which expect the class labels to be in the range [0, C-1]

Warning

This class needs scipy to load data from .mat format.

Parameters

root (string) – Root directory of dataset where directory SVHN exists.
split (string) – One of {‘train’, ‘test’, ‘extra’}. Accordingly dataset is selected. ‘extra’ is Extra training set.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where target is index of the target class.
Return type: tuple

UCF101¶

class torchvision.datasets.UCF101(root: str, annotation_path: str, frames_per_clip: int, step_between_clips: int = 1, frame_rate: Optional[int] = None, fold: int = 1, train: bool = True, transform: Optional[Callable] = None, _precomputed_metadata: Optional[Dict[str, Any]] = None, num_workers: int = 1, _video_width: int = 0, _video_height: int = 0, _video_min_dimension: int = 0, _audio_samples: int = 0)[source]¶

UCF101 dataset.

UCF101 is an action recognition video dataset. This dataset consider every video as a collection of video clips of fixed size, specified by frames_per_clip, where the step in frames between each clip is given by step_between_clips. The dataset itself can be downloaded from the dataset website; annotations that annotation_path should be pointing to can be downloaded from here <https://www.crcv.ucf.edu/data/UCF101/UCF101TrainTestSplits-RecognitionTask.zip>.

To give an example, for 2 videos with 10 and 15 frames respectively, if frames_per_clip=5 and step_between_clips=5, the dataset size will be (2 + 3) = 5, where the first two elements will come from video 1, and the next three elements from video 2. Note that we drop clips which do not have exactly frames_per_clip elements, so not all frames in a video might be present.

Internally, it uses a VideoClips object to handle clip creation.

Parameters

root (string) – Root directory of the UCF101 Dataset.
annotation_path (str) – path to the folder containing the split files; see docstring above for download instructions of these files
frames_per_clip (int) – number of frames in a clip.
step_between_clips (int, optional) – number of frames between each clip.
fold (int, optional) – which fold to use. Should be between 1 and 3.
train (bool, optional) – if True, creates a dataset from the train split, otherwise from the test split.
transform (callable, optional) – A function/transform that takes in a TxHxWxC video and returns a transformed version.

Returns

A 3-tuple with the following entries:

video (Tensor[T, H, W, C]): the T video frames

audio(Tensor[K, L]): the audio frames, where K is the number of channels and L is the number of points

label (int): class of the video clip

Return type

tuple

__getitem__(idx: int) → Tuple[torch.Tensor, torch.Tensor, int][source]¶

Parameters: index (int) – Index
Returns: Sample and meta data, optionally transformed by the respective transforms.
Return type: (Any)

USPS¶

class torchvision.datasets.USPS(root: str, train: bool = True, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

USPS Dataset. The data-format is : [label [index:value ]*256 n] * num_lines, where label lies in [1, 10]. The value for each pixel lies in [-1, 1]. Here we transform the label into [0, 9] and make pixel values in [0, 255].

Parameters

root (string) – Root directory of dataset to store``USPS`` data files.
train (bool, optional) – If True, creates dataset from usps.bz2, otherwise from usps.t.bz2.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where target is index of the target class.
Return type: tuple

VOC¶

class torchvision.datasets.VOCSegmentation(root: str, year: str = '2012', image_set: str = 'train', download: bool = False, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, transforms: Optional[Callable] = None)[source]¶

Pascal VOC Segmentation Dataset.

Parameters

root (string) – Root directory of the VOC Dataset.
year (string, optional) – The dataset year, supports years "2007" to "2012".
image_set (string, optional) – Select the image_set to use, "train", "trainval" or "val". If year=="2007", can also be "test".
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
transforms (callable, optional) – A function/transform that takes input sample and its target as entry and returns a transformed version.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where target is the image segmentation.
Return type: tuple

class torchvision.datasets.VOCDetection(root: str, year: str = '2012', image_set: str = 'train', download: bool = False, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, transforms: Optional[Callable] = None)[source]¶

Pascal VOC Detection Dataset.

Parameters

root (string) – Root directory of the VOC Dataset.
year (string, optional) – The dataset year, supports years "2007" to "2012".
image_set (string, optional) – Select the image_set to use, "train", "trainval" or "val". If year=="2007", can also be "test".
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again. (default: alphabetic indexing of VOC’s 20 classes).
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, required) – A function/transform that takes in the target and transforms it.
transforms (callable, optional) – A function/transform that takes input sample and its target as entry and returns a transformed version.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where target is a dictionary of the XML tree.
Return type: tuple

WIDERFace¶

class torchvision.datasets.WIDERFace(root: str, split: str = 'train', transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, download: bool = False)[source]¶

WIDERFace Dataset.

Parameters

root (string) –

Root directory where images and annotations are downloaded to. Expects the following folder structure if download=False:

<root>
    └── widerface
        ├── wider_face_split ('wider_face_split.zip' if compressed)
        ├── WIDER_train ('WIDER_train.zip' if compressed)
        ├── WIDER_val ('WIDER_val.zip' if compressed)
        └── WIDER_test ('WIDER_test.zip' if compressed)

split (string) – The dataset split to use. One of {train, val, test}. Defaults to train.
transform (callable, optional) – A function/transform that takes in a PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
download (bool, optional) – If true, downloads the dataset from the internet and puts it in root directory. If dataset is already downloaded, it is not downloaded again.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (image, target) where target is a dict of annotations for all faces in the image. target=None for the test split.
Return type: tuple

Base classes for custom datasets¶

class torchvision.datasets.DatasetFolder(root: str, loader: Callable[[str], Any], extensions: Optional[Tuple[str, …]] = None, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, is_valid_file: Optional[Callable[[str], bool]] = None)[source]¶

A generic data loader.

This default directory structure can be customized by overriding the find_classes() method.

Parameters

root (string) – Root directory path.
loader (callable) – A function to load a sample given its path.
extensions (tuple[string]) – A list of allowed extensions. both extensions and is_valid_file should not be passed.
transform (callable, optional) – A function/transform that takes in a sample and returns a transformed version. E.g, transforms.RandomCrop for images.
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
is_valid_file – A function that takes path of a file and check if the file is a valid file (used to check of corrupt files) both extensions and is_valid_file should not be passed.

__getitem__(index: int) → Tuple[Any, Any][source]¶

Parameters: index (int) – Index
Returns: (sample, target) where target is class_index of the target class.
Return type: tuple

find_classes(directory: str) → Tuple[List[str], Dict[str, int]][source]¶

Find the class folders in a dataset structured as follows:

directory/
├── class_x
│   ├── xxx.ext
│   ├── xxy.ext
│   └── ...
│       └── xxz.ext
└── class_y
    ├── 123.ext
    ├── nsdf3.ext
    └── ...
    └── asd932_.ext

This method can be overridden to only consider a subset of classes, or to adapt to a different dataset directory structure.

Parameters: directory (str) – Root directory path, corresponding to self.root
Raises: FileNotFoundError – If dir has no class folders.
Returns: List of all classes and dictionary mapping each class to an index.
Return type: (Tuple[List[str], Dict[str, int]])

static make_dataset(directory: str, class_to_idx: Dict[str, int], extensions: Optional[Tuple[str, …]] = None, is_valid_file: Optional[Callable[[str], bool]] = None) → List[Tuple[str, int]][source]¶

Generates a list of samples of a form (path_to_sample, class).

This can be overridden to e.g. read files from a compressed zip file instead of from the disk.

Parameters

directory (str) – root dataset directory, corresponding to self.root.
class_to_idx (Dict[str, int]) – Dictionary mapping class name to class index.
extensions (optional) – A list of allowed extensions. Either extensions or is_valid_file should be passed. Defaults to None.
is_valid_file (optional) – A function that takes path of a file and checks if the file is a valid file (used to check of corrupt files) both extensions and is_valid_file should not be passed. Defaults to None.

Raises

ValueError – In case class_to_idx is empty.
ValueError – In case extensions and is_valid_file are None or both are not None.
FileNotFoundError – In case no valid file was found for any class.

Returns

samples of a form (path_to_sample, class)

Return type

List[Tuple[str, int]]

class torchvision.datasets.ImageFolder(root: str, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None, loader: Callable[[str], Any] = <function default_loader>, is_valid_file: Optional[Callable[[str], bool]] = None)[source]¶

A generic data loader where the images are arranged in this way by default:

root/dog/xxx.png
root/dog/xxy.png
root/dog/[...]/xxz.png

root/cat/123.png
root/cat/nsdf3.png
root/cat/[...]/asd932_.png

This class inherits from DatasetFolder so the same methods can be overridden to customize the dataset.

Parameters

root (string) – Root directory path.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.
loader (callable, optional) – A function to load an image given its path.
is_valid_file – A function that takes path of an Image file and check if the file is a valid file (used to check of corrupt files)

__getitem__(index: int) → Tuple[Any, Any]¶

Parameters: index (int) – Index
Returns: (sample, target) where target is class_index of the target class.
Return type: tuple

class torchvision.datasets.VisionDataset(root: str, transforms: Optional[Callable] = None, transform: Optional[Callable] = None, target_transform: Optional[Callable] = None)[source]¶

Base Class For making datasets which are compatible with torchvision. It is necessary to override the __getitem__ and __len__ method.

Parameters

root (string) – Root directory of dataset.
transforms (callable, optional) – A function/transforms that takes in an image and a label and returns the transformed versions of both.
transform (callable, optional) – A function/transform that takes in an PIL image and returns a transformed version. E.g, transforms.RandomCrop
target_transform (callable, optional) – A function/transform that takes in the target and transforms it.

Note

transforms and the combination of transform and target_transform are mutually exclusive.

__getitem__(index: int) → Any[source]¶

Parameters: index (int) – Index
Returns: Sample and meta data, optionally transformed by the respective transforms.
Return type: (Any)

torchvision.datasets¶

Caltech¶

CelebA¶

CIFAR¶

Cityscapes¶

COCO¶

Captions¶

Detection¶

EMNIST¶

FakeData¶

Fashion-MNIST¶

Flickr¶

HMDB51¶

ImageNet¶

iNaturalist¶

Kinetics-400¶

KITTI¶

KMNIST¶

LFW¶

LSUN¶

MNIST¶

Omniglot¶

PhotoTour¶

Places365¶

QMNIST¶

SBD¶

SBU¶

SEMEION¶

STL10¶

SVHN¶

UCF101¶

USPS¶

VOC¶

WIDERFace¶

Base classes for custom datasets¶

Docs

Tutorials

Resources