GitLab

* add wide resnet

* add docstring for wide resnet

* update WRN-50-2 model

* add docs

* extend WRN docstring

* use pytorch storage for WRN

* fix rebase

* fix typo in docs

* add erase function

* add Random Erasing

* Update transforms.py

* Update transforms.py

* add test for random erasing

* Update test_transforms.py

* fix flake8

* Update test_transforms.py

* Update functional.py

* Update test_transforms.py

* fix bug for per-pixel erasing

* Update transforms.py

* specific for coordinate (x, y)

* add raise TypeError for img

* Update transforms.py

* Update transforms.rst

* Add initial mnasnet impl

* Remove all type hints, comply with PyTorch overall style

* Expose models

* Remove avgpool from features() and add separately

* Fix python3-only stuff, replace subclasses with functions

* fix __all__

* Fix typo

* Remove conditional dropout

* Make dropout functional

* Addressing @fmassa's feedback, round 1

* Replaced adaptive avgpool with mean on H and W to prevent collapsing the batch dimension

* Partially address feedback

* YAPF

* Removed redundant class vars

* Update urls to releases

* Add information to models.rst

* Replace init with kaiming_normal_ in fan-out mode

* Use load_state_dict_from_url

Values from random.random() are always in the range [0.0, 1.0), so the statement `random.random() > 5` will never be true.

* Added general reader for sn3 tensors in "pascalvincent" format

* Added class QMNIST into mnist.py

* QMNIST dataset: make some pt files smaller

* Change request from fmassa.

* read_sn3_pascalvincent_tensor: cse

* read_sn3_pascalvincent_tensor: check file size (when strict!=False)

* Fix lint

* More lint

* Add documentation and expose QMNIST to dataset namespace

* add USPS dataset

* minor fixes

* Improvements to the USPS dataset

Add it to the documentation, expose it to torchvision.datasets
and inherit from VisionDataset

Also move weights from ShuffleNet to PyTorch bucket. Additionally, rename shufflenet to make it consistent with the other models

* Add ShuffleNet v2

Added 4 configurations: x0.5, x1, x1.5, x2
Add 2 pretrained models: x0.5, x1

* fix lint

* Change globalpool to torch.mean() call

* Added Semantic Boundaries Dataset

* Updated docs and datasets importing module

* Improved docs

* Add GoogLeNet (Inception v1)

* Fix missing padding

* Add missing ReLu to aux classifier

* Add Batch normalized version of GoogLeNet

* Use ceil_mode instead of padding and initialize weights using "xavier"

* Match BVLC GoogLeNet zero initialization of classifier

* Small cleanup

* use adaptive avg pool

* adjust network to match TensorFlow

* Update url of pre-trained model and add classification results on ImageNet

* Bugfix that improves performance by 1 point

* flake_fixes

* flake_fixes2

* Add KMNIST

* Add documentation

* Add KMNIST to module's __all__ list

* Add Cityscapes Dataset

* Rename 'label' to 'semantic' to make the meaning more clear

* Add support for gtCoarse target set

* pad updated

* checked pad

* pad checked

* pad checked

* docs changed

* VOC Dataset, linted, flak8 passing, samples on gist.

* Double backtick on values.

* Apply suggestions from code review

Add suggestions from @ellisbrown, using dict of dicts instead of array index.
Co-Authored-By: bpinaya <bpg_92@hotmail.com>

* Fixed errors with the new comments.

* Added documentation on RST

* Removed getBB, added parse_voc_xml, variable naming change.

* Removed unused variable, removed VOC_CLASSES, two new gists for test.

* Add Flickr8k and Flickr30k Datasets

* Add Flickr to the dataset docs

* Sort ids, glob during construction

* annFile -> ann_file

* Fix undefined variable name bug

* Add SBU Captioned Photo Dataset

* Add SBU to the dataset docs

Give a short example of how to build a transform pipeline for a
segmentation task.

* Add LinearTransformation to docs
- Arrange the transforms alphabetically

* Add missing transforms to docs

Add testing case with Pillow-SIMD, fix bug in RandomAffine if Pillow<5.0.0, add missing docs for affine (#439)

* Adding tests to ImageFolder

* Adding DatasetFolder class

* Fix tests for pytest and code for lint checker

* Adding mock to requirements for ImageFolder tests

* Remove mocks from requirements

This reverts commit ab03dc43.

* Adding tests to ImageFolder

* Adding DatasetFolder class

* Fix tests for pytest and code for lint checker

* Adding mock to requirements for ImageFolder tests

It was pointing to an old version where the validation pre-processing was the same as the training pre-processing, and thus had randomness in it (instead of CenterCrop).

* EMNIST dataset + speedup *MNIST preprocessing

* add docs for grayscale + random grayscale transforms