GitLab

Fixes https://github.com/pytorch/vision/issues/397

* Add basic Omniglot dataset loader

* Remove unused import

* Add Omniglot random pair to sample pair of characters

* Precompute random set of pairs, deterministic after object instantiation

* Export OmniglotRandomPair via the datasets module interfact

* Fix naming convention, use sum instead of reduce

* Fix downloading to not download everything, fix Python2 syntax

* Fix end line lint

* Add random_seed, syntax fixes

* Remove randomized pair, take up as a separate generic wrapper

* Fix master conflict

* add __repr__ for datasets

* fix lint

* add __repr__ for all transform objects

* add tests for __repr__

* Update densenet.py

added missing weight init for densenet

* Update densenet.py

following official implementation [here](https://github.com/liuzhuang13/DenseNet/blob/master/models/densenet.lua#L160)

* Update densenet.py

changed to nn.init & fixed lint

* add scale and ratio params to RandomResizedCrop.get_params

* don't use Image as a context manager to avoid exceptions on close

* flake8

* EMNIST dataset + speedup *MNIST preprocessing

* param to cause FakeData to generate different images

docstring fix for Normalize: the std sequence should be (S1, ..., Sn).

* add tunable crop scale and aspect ratio in

* Added urls to Harris patches on Phototour

* fixed formatting

* fixed flake8 complains

* add docs for grayscale + random grayscale transforms

* add to_grayscale + randomGrayscale

add rotate and RandomRotation to transforms

* First commit for semeion dataset

SEMEION Handwritten Digits Data Set
http://archive.ics.uci.edu/ml/datasets/semeion+handwritten+digit

* SEMEION Class

* Fix Lint errors

* More linting errors

is_image_file is now case insensitive
Added docstring to is_image_file
is_image_file now returns true for pgm files

added CIFAR100 autoclass in docs

* refactor transforms and add transforms/functional.py

* add __all__ to transforms.py

* add docs to FiveCrop + TenCrop

* fix typo in docstring

* add assert for byte tensor to_pil_image

* add mode param to ToPILImage

* add some more tests

* fix formatiing on error messages

* added tests for to_pil_image with mode param

* flake8 fixes

* added Whiten transform and relative section in README

* added Whiten transform and its section in README

* fixed lint

* fixed lint again

* added linear transformation and whitening test unitary covariance

* fixed lint and test

* commit before rebasing

* before rebase

* added assertion to check matrix size

* fix readme

* added space around operator

* replace assert with raise ValueError, fix spaces and readme

PIL.Image is renamed to 'PIL Image' because PIL.Image is the module and PIL.Image.Image is the class we are referring but is too long to put in docstrings

Pattern from: https://packaging.python.org/guides/single-sourcing-package-version/

* Add adjust_hue and adjust_saturation

* Add adjust_brightness, adjust_contrast

Also
* Change adjust_saturation to use pillow implementation
* Documentation made clear

* Add adjust_gamma

* Add ColorJitter

* Address review comments

* Fix documentation for ColorJitter

* Address review comments 2

* Fallback to adjust_hue in case of BW images

* Add tests

* fix dtype