GitLab

* Minor refactoring of ShuffleNetV2

Added progress flag following #875. Further the following refactoring was also done:

1) added `version` argument in shufflenetv2 method and removed the operations for converting the `width_mult` arg to float and string.
2) removed `num_classes` argument and **kwargs from functions except `ShuffleNetV2`

* removed `version` arg

* Update shufflenetv2.py

* Removed the try except block

* Update shufflenetv2.py

* Changed version from float to str

* Replace `width_mult` with `stages_out_channels`

Removes the need of  `_getStages` function.

* remove 'input_size' parameter from shufflenetv2

* Update shufflenetv2.py

* 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 progress flag to model getters

* flake8

* bug fix

* backward commpability

* [RFC] Add support for joint transformations in VisionDataset

* Add joints transforms for VOC and SBD

Breaking change in SBD, the xy_transform has been renamed transforms. I think this is fine given that we have not released a version of torchvision that contains it

* final changes

* final

* linter

* test changes

* linter

* lint

* indent

* lint

* minor changes

* parameter added

* ci

* ci fixes

* indent

* indent

* indent

* arg fixed

* fixed check integrity

* stylistic changes

* flake8

* added is_valid_file option

* small fixes

* fixes

* flake8 fixes

* some test

* flake8 fixes

* improvements

* modifications on tests
* fixes

* minor fix

* Add dilation option to ResNet

* Add a size check for replace_stride_with_dilation

* Fix wrong doc string

* bug fix

* Fix ResNeXt model defs with backwards compat for ResNet.

* Fix Python 2.x integer div issue

* fix inconsistent behavior for ~ expression

* remove duplicated code

* Update test_transforms.py

* Update transforms.py

* introduced new super class for all vision datasets

* Removed root from repr if dataset has no root

* reverted some overly-ambitious autoformatting

* reverted some overly-ambitious autoformatting

* added split attribute to repr of STL10 dataset

* made Python2 friendly and more robust

* Fixed call of the superclass constructor

* moved transform and target_transform back to the base classes

* added check if transforms are present before printing to avoid setting them within the constructor

* added missing transforms and target_transforms to base classes

* fixed linter error

* add aux_logits support to inception

it is related to pytorch/pytorch#18668

* instantiate InceptionAux only when requested

it is related to pytorch/pytorch#18668

* revert googlenet

* support and aux_logits in pretrained models

* return namedtuple when aux_logit is True

str.endswith can take a single string, or a tuple of strings. This replaces a loop over each extension and a call to endswith with a single call to endswith passing in all the allowed extensions. This has the advantage that the loop over each extension is done in c rather than python, and the code is a little less verbose.

* Add groups support to ResNet

* Kill BaseResNet

* Make it support multi-machine training

Making references/classification/train.py and references/classification/utils.py  compatible with python2  (#831)

* linter fixes

* linter fixes

* remove duplicate code from densenet

* correct indentation

* Match Tensorflows implementation of GoogLeNet

* just disable the branch when pretrained is true

* don't use legacy code

Changed 'channely' to 'channel' in the comments of normalize function.