Differences with torchvision.datasets.ImageFolder

torchvision.datasets.ImageFolder

class torchvision.datasets.ImageFolder(
    root: str,
    transform: Optional[Callable] = None,
    target_transform: Union[Callable, NoneType] = None,
    loader: Optional[Callable] = None,
    is_valid_file: bool = None)

For more information, see torchvision.datasets.ImageFolder.

mindspore.dataset.ImageFolderDataset

class mindspore.dataset.ImageFolderDataset(
    dataset_dir,
    num_samples=None,
    num_parallel_workers=None,
    shuffle=None,
    sampler=None,
    extensions=None,
    class_indexing=None,
    decode=False,
    num_shards=None,
    shard_id=None,
    cache=None,
    decrypt=None)

For more information, see mindspore.dataset.ImageFolderDataset.

Differences

PyTorch: A source dataset that reads images from a tree of directories. API integrates the transformation operations for image and label. File Loader can be specified.

MindSpore: A source dataset that reads images from a tree of directories. Transforms for image and label depends on mindshare.dataset.map operation. File Loader can not be specified.

Categories	Subcategories	PyTorch	MindSpore	Difference
Parameter	Parameter1	root	dataset_dir	-
	Parameter2	transform	-	Supported by mindspore.dataset.map operation
	Parameter3	target_transform	-	Supported by mindspore.dataset.map operation
	Parameter4	loader	-	Not supported by MindSpore
	Parameter5	is_valid_file	-	Not supported by MindSpore
	Parameter6	-	num_samples	The number of images to be included in the dataset
	Parameter7	-	num_parallel_workers	Number of worker threads to read the data
	Parameter8	-	shuffle	Whether to perform shuffle on the dataset
	Parameter9	-	sampler	Object used to choose samples from the dataset
	Parameter10	-	extensions	List of file extensions to be included in the dataset
	Parameter11	-	class_indexing	A str-to-int mapping from folder name to index
	Parameter12	-	decode	Whether to decode the images after reading
	Parameter13	-	num_shards	Number of shards that the dataset will be divided into
	Parameter14	-	shard_id	The shard ID within num_shards
	Parameter15	-	cache	Use tensor caching service to speed up dataset processing
	Parameter16	-	decrypt	Image decryption function

Code Example

Assume that we have a directory with the following structure：

imageset/
    ├── cat
    │   ├── cat_0.jpg
    │   ├── cat_1.jpg
    │   └── cat_2.jpg
    ├── fish
    │   ├── fish_0.jpg
    │   ├── fish_1.jpg
    │   ├── fish_2.jpg
    │   └── fish_3.jpg
    ├── fruits
    │   ├── fruits_0.jpg
    │   ├── fruits_1.jpg
    │   └── fruits_2.jpg
    ├── plane
    │   ├── plane_0.jpg
    │   ├── plane_1.jpg
    │   └── plane_2.jpg
    └── tree
        ├── tree_0.jpg
        ├── tree_1.jpg
        └── tree_2.jpg

# Torch
import torchvision.transforms as T
import torchvision.datasets as datasets
from torch.utils.data import DataLoader

root = "/path/to/imageset/"
dataset = datasets.ImageFolder(root, transform=T.RandomCrop((256, 256)))
dataloader = DataLoader(dataset)

# MindSpore
import mindspore.dataset as ds
import mindspore.dataset.vision as vision

root = "/path/to/imageset/"
ms_dataloader = ds.ImageFolderDataset(root, decode=True)
ms_dataloader = ms_dataloader.map(vision.RandomCrop((256, 256)), ["image"])