GitLab

* [Remove] Use stride in 1x1 in resnet

This is temporary

* Move files to torchvision

Inference works

* Now seems to give same results

Was using the wrong number of total iterations in the end...

* Distributed evaluation seems to work

* Factor out transforms into its own file

* Enabling horizontal flips

* MultiStepLR and preparing for launches

* Add warmup

* Clip gt boxes to images

Seems to be crucial to avoid divergence. Also reduces the losses over different processes for better logging

* Single-GPU batch-size 1 of CocoEvaluator works

* Multi-GPU CocoEvaluator works

Gives the exact same results as the other one, and also supports batch size > 1

* Silence prints from pycocotools

* Commenting unneeded code for run

* Fixes

* Improvements and cleanups

* Remove scales from Pooler

It was not a free parameter, and depended only on the feature map dimensions

* Cleanups

* More cleanups

* Add misc ops and totally remove maskrcnn_benchmark

* nit

* Move Pooler to ops

* Make FPN slightly more generic

* Minor improvements or FPN

* Move FPN to ops

* Move functions to utils

* Lint fixes

* More lint

* Minor cleanups

* Add FasterRCNN

* Remove modifications to resnet

* Fixes for Python2

* More lint fixes

* Add aspect ratio grouping

* Move functions around

* Make evaluation use all images for mAP, even those without annotations

* Bugfix with DDP introduced in last commit

* [Check] Remove category mapping

* Lint

* Make GroupedBatchSampler prioritize largest clusters in the end of iteration

* Bugfix for selecting the iou_types during evaluation

Also switch to using the torchvision normalization now on, given that we are using torchvision base models

* More lint

* Add barrier after init_process_group

Better be safe than sorry

* Make evaluation only use one CPU thread per process

When doing multi-gpu evaluation, paste_masks_in_image is multithreaded and throttles evaluation altogether. Also change default for aspect ratio group to match Detectron

* Fix bug in GroupedBatchSampler

After the first epoch, the number of batch elements could be larger than batch_size, because they got accumulated from the previous iteration. Fix this and also rename some variables for more clarity

* Start adding KeypointRCNN

Currently runs and perform inference, need to do full training

* Remove use of opencv in keypoint inference

PyTorch 1.1 adds support for bicubic interpolation which matches opencv (except for empty boxes, where one of the dimensions is 1, but that's fine)

* Remove Masker

Towards having mask postprocessing done inside the model

* Bugfixes in previous change plus cleanups

* Preparing to run keypoint training

* Zero initialize bias for mask heads

* Minor improvements on print

* Towards moving resize to model

Also remove class mapping specific to COCO

* Remove zero init in bias for mask head

Checking if it decreased accuracy

* [CHECK] See if this change brings back expected accuracy

* Cleanups on model and training script

* Remove BatchCollator

* Some cleanups in coco_eval

* Move postprocess to transform

* Revert back scaling and start adding conversion to coco api

The scaling didn't seem to matter

* Use decorator instead of context manager in evaluate

* Move training and evaluation functions to a separate file

Also adds support for obtaining a coco API object from our dataset

* Remove unused code

* Update location of lr_scheduler

Its behavior has changed in PyTorch 1.1

* Remove debug code

* Typo

* Bugfix

* Move image normalization to model

* Remove legacy tensor constructors

Also move away from Int and instead use int64

* Bugfix in MultiscaleRoiAlign

* Move transforms to its own file

* Add missing file

* Lint

* More lint

* Add some basic test for detection models

* More lint

* Initial version of the segmentation examples

WIP

* Cleanups

* [WIP]

* Tag where runs are being executed

* Minor additions

* Update model with new resnet API

* [WIP] Using torchvision datasets

* Improving datasets

Leverage more and more torchvision datasets

* Reorganizing datasets

* PEP8

* No more SegmentationModel

Also remove outplanes from ResNet, and add a function for querying intermediate outputs. I won't keep it in the end, because it's very hacky and don't work with tracing

* Minor cleanups

* Moving transforms to its own file

* Move models to torchvision

* Bugfixes

* Multiply LR by 10 for classifier

* Remove classifier x 10

* Add tests for segmentation models

* Update with latest utils from classification

* Lint and missing import

* Initial layout for layers with cpp extensions

* Move files around

* Fix import after move

* Add support for multiple types to ROIAlign

* Different organization

CUDA extensions work now

* Cleanups

* Reduce memory requirements for backwards

* Replace runtime_error by AT_ERROR

* Add nms test

* Add support for compilation using CPP extensions

* Change folder structure

* Add ROIPool cuda

* Cleanups

* Add roi_pool.py

* Fix lint

* Add initial structures folder for bounding boxes

* Assertion macros compatible with pytorch master (#540)

* Support for ROI Pooling (#592)

* ROI Pooling with tests. Fix for cuda context in ROI Align.

* renamed bottom and top to follow torch conventions

* remove .type().tensor() calls in favor of the new approach to tensor initialization (#626)

* Consistent naming for rois variable (#627)

* remove .type().tensor() calls in favor of the new approach to tensor initialization

* Consistent naming for rois variable in ROIPool

* ROIPool: Support for all datatypes (#632)

* Use of torch7 naming scheme for ROIAlign forward and backward

* use common cuda helpers in ROIAlign

* use .options() in favor of .type() where applicable

* Added tests for forward pass of ROIAlign, as well as more consistent naming scheme for CPU vs CUDA

* working ROIAlign cuda backwards pass

* working ROIAlign backwards pass for CPU

* added relevant headers for ROIAlign backwards

* tests for ROIAlign layer

* replace .type() with .options() for tensor initialization in ROIAlign layers

* support for Half types in ROIAlign

* gradcheck tests for ROIAlign

* updated ROIPool on CPU to work with all datatypes

* updated and cleaned tests for ROI Pooling

* Fix rebase problem

* Remove structures folder

* Improve cleanup and bugfix in test_layers

* Update C++ headers

* Add CUDAGuard to cu files

* Add more checks to layers

* Add CUDA NMS and tests

* Add multi-type support for NMS CUDA

* Avoid using THCudaMalloc

* Add clang-format and reformat c++ code

* Remove THC includes

* Rename layers to ops

* Add documentation and rename functions

* Improve the documentation a bit

* Fix some lint errors

* Fix remaining lint inssues

* Area computation doesn't add +1 in NMS

* Update CI to use PyTorch nightly

* Make NMS return indices sorted according to the score

* Address reviewer comments

* Lint fixes

* Improve doc for roi_align and roi_pool

* move to xenial

* Fix bug pointed by @lopuhin

* Fix RoIPool reference implementation in Python 2

Also fixes a bug in the clip_boxes_to_image -- this function needs a test!

* Remove change in .travis

* fixed check integrity

* stylistic changes

* added test for check_md5 and check_integrity

* flake8

* fix path to test file if not executed from test folder

* final changes

* final

* linter

* test changes

* linter

* lint

* indent

* lint

* minor changes

* parameter added

* ci

* ci fixes

* indent

* indent

* indent

* arg fixed

* 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

* Update test_transforms.py

* Update transforms.py

* Add test for loading pretrained models

The update modifies the test to check whether the model can successfully load the pretrained weights. Will raise an error if the model parameters are incorrectly defined or named.

* Add test on 'num_class'

Passing num_class equal to a number other than 1000 helps in making the test more enforcing in nature.

* Add basic model testing.

Also fixes flaky test

* Fix flake8

* Add Affinetransformation

Add Affinetransformation to superseed LinearTransformation

* Add test

* Add zero mean_vector in LinearTransformation and improved docs

* update

* minor fix

* minor fix2

* fixed flake8

* fix flake8

* fixed transpose syntax

* fixed shape of mean_vector in test

* fixed test

* print est cov and mean

* fixed flake8

* debug

* reduce num_samples

* debug

* fixed num_features

* fixed rtol for cov

* fix __repr__

* Update transforms.py

* Update test_transforms.py

* Update transforms.py

* fix flake8

* Update transforms.py

* Update transforms.py

* Update transforms.py

* Update transforms.py

* Changed dim of mean_vector to 1D, doc and removed .numpy () from format_string

* Restore test_linear_transformation()

* Update test_transforms.py

* Fixes for PyTorch version of tqdm

* Flake

* Flake fix

* randomresizedmods

* lint checks

* test to randomrescrop added

* updates

* tests updated

* tests updated

* upd

* updates

* Update torchvision/transforms/transforms.py
Co-Authored-By: surgan12 <33121121+surgan12@users.noreply.github.com>

* tests changed

* trvis

* travis

* fixes syntax

* ...

* flake fixes

* flake_fixes

* flake_fixes2

The tests were previously taking 2 minutes, not they take 4 seconds

* modes added

* tests_added

* Update test_transforms.py

* Update test_transforms.py

* Update test_transforms.py

* added separate checks for dimensionality in to_pil_image and added tests

* updated to_pil_image to use both 2D ndarrays and tensors, as well as refactored the tests

* Revert "move area calculation out of loop (#641)"

This reverts commit 62cbf0bf.

* Revert "Fix FashionMNIST loading MNIST (#640)"

This reverts commit 36215690.

Before this would lead FashionMNIST to contain mnist data:

```
MNIST(root, download=True)
FashionMNIST(root, download=True)
```

As MNIST and FashionMNIST are the same classes, the processed
outputs actual ended up to be the same files. This commit now
stores them at different files and also stores the class name when
saving them. I also added md5 sums.

* cast mnist targer to int

* fix unused variables

* fix syntax in tests

* remove # noqa; rename l variable to line_split; use except OSError

* add W504

* add W504

* add W503,W504 tox.ini

Improve `test_pad_with_non_constant_padding_modes` to avoid data multiplication in `transforms.ToPILImage()` on float data:
```python
img = torch.zeros(3, 27, 27)  # Float32
img[:, :, 0] = 1   # we add 1 and not 255
img = transforms.ToPILImage()(img)  # This converts 1 to 255 due to [pic = pic.mul(255).byte()](https://github.com/pytorch/vision/blob/master/torchvision/transforms/functional.py#L107)
```
and thus test's correct values are [..., 200, 1, 0]

Remove test with variables

* Add case in test_to_tensor for PIL Images mode '1'

* Add support in ToTensor for PIL Images mode '1'

* Fix pep8 issues

* Made transorms.RandomCrop tolerate images smaller than the given size.

* Extended the tescase for transforms.RandomCrop

* Made the testcase test for owidth=width+1

* Fixed the one pixel pading and the testcase

* Fixed minor lintint errors. flake8 passes.

* Added reflect, symmetric and edge padding

* Updated padding docs, added tests

* 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

* Add random affine transformation

* Rewrite __repr__

* Refactor affine transform and update tests

* Add RandomApply, RandomChoice, RandomOrder transformations

* Rename argument `proba` to `p`

* Set probability as configuration parameter in RandomHorizontalFlip and RandomHorizontalFlip (#414)

* Fix documentation