Datasets

This subpackage can be used for generating reconstruction datasets from other base datasets (e.g. MNIST or CelebA).

HD5Dataset

`deepinv.datasets.HDF5Dataset`	DeepInverse HDF5 dataset with signal/measurement pairs `(x, y)`.
`deepinv.datasets.generate_dataset`	Generates dataset of signal/measurement pairs from base dataset.

Generating a dataset associated with a certain forward operator is done via deepinv.datasets.generate_dataset() using a base PyTorch dataset (torch.utils.data.Dataset, in this case MNIST). For example, here we generate a compressed sensing MNIST dataset:

Note

We support all data types supported by h5py, including complex numbers.

>>> import deepinv as dinv
>>> from torchvision import datasets, transforms
>>>
>>> save_dir = '../datasets/MNIST/' # directory where the dataset will be saved.
>>>
>>> # define base train dataset
>>> transform_data = transforms.Compose([transforms.ToTensor()])
>>> data_train = datasets.MNIST(root='../datasets/', train=True,
...                             transform=transform_data, download=True)
>>> data_test = datasets.MNIST(root='../datasets/', train=False, transform=transform_data)
>>>
>>> # define forward operator
>>> physics = dinv.physics.CompressedSensing(m=300, img_shape=(1, 28, 28))
>>> physics.noise_model = dinv.physics.GaussianNoise(sigma=.05)
>>>
>>> # generate paired dataset
>>> generated_dataset_path = dinv.datasets.generate_dataset(train_dataset=data_train, test_dataset=data_test,
...                                physics=physics, save_dir=save_dir, verbose=False)

Similarly, we can generate a dataset from a local folder of images (other types of data can be loaded using the loader and is_valid_file arguments of torchvision.datasets.ImageFolder`()):

>>> # Note that ImageFolder requires file structure to be '.../dir/train/xxx/yyy.ext' where xxx is an arbitrary class label
>>> data_train = datasets.ImageFolder(f'{save_dir}/train', transform=transform_data)
>>> data_test  = datasets.ImageFolder(f'{save_dir}/test',  transform=transform_data)
>>>
>>> dinv.datasets.generate_dataset(train_dataset=data_train, test_dataset=data_test,
>>>                                physics=physics, device=dinv.device, save_dir=save_dir)

The datasets are saved in .h5 (HDF5) format, and can be easily loaded to pytorch’s standard torch.utils.data.DataLoader:

>>> from torch.utils.data import DataLoader
>>>
>>> dataset = dinv.datasets.HDF5Dataset(path=generated_dataset_path, train=True)
>>> dataloader = DataLoader(dataset, batch_size=4, shuffle=True)

We can also use physics generators to randomly generate physics params for data, and save and load the physics params into the dataset:

>>> physics_generator = dinv.physics.generator.SigmaGenerator()
>>> pth = dinv.datasets.generate_dataset(train_dataset=data_train, test_dataset=data_test,
>>>                                      physics=physics, physics_generator=physics_generator,
>>>                                      device=dinv.device, save_dir=save_dir)
>>> dataset = dinv.datasets.HDF5Dataset(path=pth, load_physics_generator_params=True, train=True)
>>> dataloader = DataLoader(dataset, batch_size=4, shuffle=True)
>>> x, y, params = next(iter(dataloader))
>>> print(params['sigma'].shape)
torch.Size([4])

PatchDataset

Generate a dataset of all patches out of a tensor of images.

deepinv.datasets.PatchDataset

Builds the dataset of all patches from a tensor of images.

Image Datasets

Ready-made datasets available in the deepinv.datasets module.

`deepinv.datasets.DIV2K`	Dataset for DIV2K Image Super-Resolution Challenge.
`deepinv.datasets.Urban100HR`	Dataset for Urban100.
`deepinv.datasets.Set14HR`	Dataset for Set14.
`deepinv.datasets.CBSD68`	Dataset for CBSBD68.
`deepinv.datasets.FastMRISliceDataset`	Dataset for fastMRI that provides access to MR image slices.
`deepinv.datasets.LidcIdriSliceDataset`	Dataset for LIDC-IDRI that provides access to CT image slices.
`deepinv.datasets.Flickr2kHR`	Dataset for Flickr2K.
`deepinv.datasets.LsdirHR`	Dataset for LSDIR.
`deepinv.datasets.FMD`	Dataset for Fluorescence Microscopy Denoising.