GitLab

* add autocasting on ROCm

* enable ROIAlign autocasting on ROCm

* enable NMS autocasting on ROCm

* fix to use correct torch CUDA APIs

* Added cuda torch.jit tests for rgb-hsv tensor conversion ops

* Reverted deleted test

* [WIP] Unify ops Grayscale and RandomGrayscale

* Unified inputs for grayscale op and transforms
- deprecated F.to_grayscale in favor of F.rgb_to_grayscale

* Fixes bug with fp input

* Rewritten adjust_* tests
- split test_adjustments into 3 separate tests
- unified testing approach with test_adjust_gamma

* Added ColorJitter tests

* Relaxed tolerance for functional adjust-* tests

* Removed wrong merge and commented code

Upstream is on 9.4.1 and we were experiencing issues when conda-build
was attempting to check for overlinking and failed out due to some files
not existing in xcode 9.0
Signed-off-by: Eli Uriegas <eliuriegas@fb.com>

* [WIP] Unify ops Grayscale and RandomGrayscale

* Unified inputs for grayscale op and transforms
- deprecated F.to_grayscale in favor of F.rgb_to_grayscale

* Fixes bug with fp input

* [WIP] Updated code according to review

* Removed unused import

* Add support for export ROIAlign

* Fix for feedback

* flake8

* fix images extraction

* remove test split

* fix tests

* be less clever in test data generation

* remove micro optimization

* lint

* Improve transforms test codebase
- refactored compareTensorToPIL, _create_data, approxEqualTensorToPIL methods

* Fixed flake8

* initial draft

* [dirty] progress

* remove inheritance from ImageFolder

* add tests

* lint

* fix type hints

* align getitem with other datasets

* remove unused import

* add docstring

* guard existing image folders from overwrite

* add missing entry in docstring

* make fixpath more legible

* add Places365 to docs

* [WIP] Fixes #2598
- Adjusted num_kernels and batch_size according to kMaxGridNum * CUDA_NUM_THREADS
- Tests to add

* - Redefined kMaxGridNum as max grid according to current CUDA device
- Added test to check the code from issue and compared grads CPU/CUDA

* Fixed static kMaxGridNum evaluation to dynamic

* Update README.rst

Updated the link to the docs..

* Update README.rst

- can be helpful for debugging C++ extensions

offset's 1 dimension should be batch size

* Adapted almost all functional tensor tests on CPU/CUDA
- fixed bug with transforms using generated grid
- remains *_crop, blocked by #2568
- TODO: test_adjustments

* Apply suggestions from code review
Co-authored-by: Francisco Massa <fvsmassa@gmail.com>

* Fixed issues according to review

* Split tests into two: cpu and cuda

* Updated test_adjustments to run on CPU and CUDA
Co-authored-by: Francisco Massa <fvsmassa@gmail.com>

* Only pull keys from db in lsun for faster cache.

This pull request inhances the speed of the cache creation for lsun dataset. For the "kitchen_train" the speed was getting slow with cache creation taking more then two hours. This speeds up to cache creation in within minutes. The issue was pulling the large image values each time and dropping them.

For more details on this please refer this issue https://github.com/jnwatson/py-lmdb/issues/195.

* Fixed bug in lsun.py when loading multiple categories

* Make linter happy

* Upgrade ORT to fix build

* Update .travis.yml

- Updated docs
- Put warning in the code
- Updated tests

* Unified input for RandomPerspective

* Updated docs

* Fixed failing test and bug with torch.randint

* Update test_functional_tensor.py

* [WIP] Added unified input perspective transformation code

* Unified input for F.perspective
- added tests
- updated docs

* Added more random test configs

* Fixed the code according to PR's review

* Added code for F_t.rotate with test
- updated F.affine tests

* Rotate test tolerance to 2%

* Fixes failing test

* [WIP] RandomRotation

* Unified RandomRotation with tests

* Added code for F_t.rotate with test
- updated F.affine tests

* Rotate test tolerance to 2%

* Fixes failing test

* Optimized _expanded_affine_grid with a single matmul op

* Recoded _compute_output_size

* [WIP] recoded F_t.rotate internal methods

* [WIP] Fixed F.affine to support rectangular images

* Recoded _gen_affine_grid to optimized version ~ affine_grid
- Fixes flake8

* [WIP] Use _gen_affine_grid for affine and rotate

* Fixed tests on square / rectangular images for affine and rotate ops

* Removed redefinition of F.rotate
- due to bad merge

* Added code for F_t.rotate with test
- updated F.affine tests

* Rotate test tolerance to 2%

* Fixes failing test

* Optimized _expanded_affine_grid with a single matmul op

* Recoded _compute_output_size

* Add fake data generator for UCF101

* Minor error correction

* Reduce total number of categories

* Fix naming

* Increase length

* Store in uint8

* Close fds

* Add assertGreater

* Add dimension tests

* Use numel instead of size

* Iterate over folds and splits

* fix type hints and move degenerate boxes to a function in torchvision.models.detection.generalized_rcnn

* format code

* format code

* changed to static method

* revert imports

* changed to method

* revert procedure for degenerating boxes

* Fixed resize to only match the smaller edge when an int is given instead of always

* Added test cases for resize()

* add typehints for torchvision.datasets.stl10

* move annotation from class to instance scope

* Clarify scaling of ToTensor transform for target masks in the documentation

* Add link to the segmentation reference
Co-authored-by: vfdev <vfdev.5@gmail.com>
Co-authored-by: vfdev <vfdev.5@gmail.com>