GitLab

* F.normalize unsqueeze mean & std if necessary

* added tests to F.normalize for 3d mean & std tensors

* modified FasterRCNN to accept negative samples

* remove debug lines

* Change torch.zeros_like to torch.zerros

* Add unit tests

* take the `device` into account
Co-authored-by: Francisco Massa <fvsmassa@gmail.com>

Co-authored-by: Francisco Massa <fvsmassa@gmail.com>

* Base decoder for video. (#1747)

Summary:
Pull Request resolved: https://github.com/pytorch/vision/pull/1747

Pull Request resolved: https://github.com/pytorch/vision/pull/1746

Added the implementation of ffmpeg based decoder with functionality that can be used in VUE and TorchVision.

Reviewed By: fmassa

Differential Revision: D19358914

fbshipit-source-id: abb672f89bfaca6351dda2354f0d35cf8e47fa0f

* Integrated base decoder into VideoReader class and video_utils.py (#1766)

Summary:
Pull Request resolved: https://github.com/pytorch/vision/pull/1766

Replaced FfmpegDecoder (incompativle with VUE) by base decoder (compatible with VUE).
Modified python utilities video_utils.py for internal simplification. Public interface got preserved.

Reviewed By: fmassa

Differential Revision: D19415903

fbshipit-source-id: 4d7a0158bd77bac0a18732fe4183fdd9a57f6402

* Optimizating base decoder performance. (#1852)

Summary:
Pull Request resolved: https://github.com/pytorch/vision/pull/1852

Changed base decoder internals for a faster clip processing.

Reviewed By: stephenyan1231

Differential Revision: D19748379

fbshipit-source-id: 58a435f0a0b25545e7bd1a3edb0b1d558176a806

* Minor fix and decoder class members access.

Summary:
Found and fix a bug in cropping algorithm (simple mistyping).
Also derived classes need access to some decoder class members, like initialization parameters - make it protected.

Reviewed By: stephenyan1231, fmassa

Differential Revision: D19895076

fbshipit-source-id: 691336c8e18526b085ae5792ac3546bc387a6db9

* Added missing header for less dependencies. (#1898)

Summary:
Pull Request resolved: https://github.com/pytorch/vision/pull/1898

Include streams/samplers shouldn't depend on decoder headers. Add dependencies directly to the place where they are required.

Reviewed By: stephenyan1231

Differential Revision: D19911404

fbshipit-source-id: ef322a053708405c02cee4562b456b1602fb12fc

* Implemented VUE Asynchronous Decoder

Summary: For Mothership we have found that asynchronous decoder provides a better performance.

Differential Revision: D20026194

fbshipit-source-id: 627b91844b4e3f917002031dd32cb19c239f4ba8

* fix a bug in API read_video_from_memory (#1942)

Summary:
Pull Request resolved: https://github.com/pytorch/vision/pull/1942

In D18720474, it introduces a bug in `read_video_from_memory` API. Thank weiyaowang for reporting it.

Reviewed By: weiyaowang

Differential Revision: D20270179

fbshipit-source-id: 66348c99a5ad1f9129b90e934524ddfaad59de03

* extend decoder to support new video_max_dimension argument (#1924)

Summary:
Pull Request resolved: https://github.com/pytorch/vision/pull/1924

Extend `video reader` decoder python API in Torchvision to support a new argument `video_max_dimension`. This enables the new video decoding use cases. When setting `video_width=0`, `video_height=0`, `video_min_dimension != 0`, and `video_max_dimension != 0`, we can rescale the video clips so that its spatial resolution (height, width) becomes
 - (video_min_dimension, video_max_dimension) if original height < original width
 - (video_max_dimension, video_min_dimension) if original height >= original width

This is useful at video model testing stage, where we perform fully convolution evaluation and take entire video frames without cropping as input. Previously, for instance we can only set `video_width=0`, `video_height=0`, `video_min_dimension = 128`, which will preserve aspect ratio. In production dataset, there are a small number of videos where aspect ratio is either extremely large or small, and when the shorter edge is rescaled to 128, the longer edge is still large. This will easily cause GPU memory OOM when we sample multiple video clips, and put them in a single minibatch.

Now, we can set (for instance) `video_width=0`, `video_height=0`, `video_min_dimension = 128` and `video_max_dimension = 171` so that the rescale resolution is either (128, 171) or (171, 128) depending on whether original height is larger than original width. Thus, we are less likely to have gpu OOM because the spatial size of video clips is determined.

Reviewed By: putivsky

Differential Revision: D20182529

fbshipit-source-id: f9c40afb7590e7c45e6908946597141efa35f57c

* Fixing samplers initialization (#1967)

Summary:
Pull Request resolved: https://github.com/pytorch/vision/pull/1967



No-ops for torchvision diff, which fixes samplers.

Differential Revision: D20397218

fbshipit-source-id: 6dc4d04364f305fbda7ca4f67a25ceecd73d0f20

* Exclude C++ test files
Co-authored-by: Yuri Putivsky <yuri@fb.com>
Co-authored-by: Zhicheng Yan <zyan3@fb.com>

Co-authored-by: eellison <eellison@fb.com>

Co-authored-by: Charles Pao <dirtybluer@gmail.com>

* add checkout/assert in roi_pool

* add checkout/assert in roi_align

* move check_roi_boxes_shape func to ops/_utils.py

* add tests

* fix CI

* fix CI
Co-authored-by: Guanheng Zhang <zhangguanheng@devfair0197.h2.fair>

https://github.com/pytorch/vision/pull/1949 seems to forget fixing quantized googlenet

* update model path

* remove autologits before loading quantized model

* Added code to support creating extension on ROCm

* max -> fmaxf conversion for hipification

* added WITH_HIP flag for hipExtension

* added appropriate headers for HIP build

* use USE_ROCM in condition to build

* change fmaxf and fminf calls

* fminf -> min

* fix the check for ROCM_HOME

* more robust checking for rocm pytorch

* add check for pytorch version before using HIP extensions

* conditional reading of ROCM_HOME

* fix googlenet no aux logits

* small fix
Co-authored-by: eellison <eellison@fb.com>

* Aligned flag in the interfaces

* Aligned flag in the impl, and remove unused comments

* Handling empty bin in forward

* Remove raise error in roi_width

* Aligned flag in the Testcodes

* Improved error messages for transforms.functional.normalize().

Split the original TypeError into 1. a TypeError if `tensor` is not a torch.Tensor and 2. a ValueError if `tensor` does not have the correct dimensionality.

Added more detail to the error for when `tensor` has the wrong dimension to make it easier to diagnose. This is useful when this function isn't called directly by the user (e.g. when the user uses transforms.Normalize and can't directly see this functions doc string).

Deleted hanging function `_is_tensor_image()`. It isn't used in this file and isn't used internally anywhere else in torchvision that I can see. (Some users will have used it despite the underscore prefix, but a quick google search for "F._is_tensor_image" suggests this is rare).

* Value checking to prevent division by zero runtime crashes.

Added a ValueError to check for and avoid division by zero in `div_`. Not preventing the call leads to runtime crashes, at least in some environments.

* Fixed div by zero check for non-scalar inputs.

* docs for faster+mask rcnn coords is clearer

* keypoint rcnn coords format is clearer
Co-authored-by: rvirgolireply <51229032+rvirgolireply@users.noreply.github.com>

Co-authored-by: rvirgolireply <51229032+rvirgolireply@users.noreply.github.com>

* replaced mean on dimensions 2,3 by adaptive_avg_pooling2d with destination of 1, to remove hardcoded dimension ordering

* replaced reshape command by torch.squeeze after global_avg_pool2d, which is cleaner

* reshape rather than squeeze for BS=1

* remove import torch

input to parameter mode can be 'fine' or 'coarse'. The code internally converts the mode = 'fine' input to 'gtFine' (same for coarse input and gtCoarse), but the docs mention that input to mode should be 'gtFine' or 'gtCoarse' inconsistently.

* feat: Added __repr__ attribute to GeneralizedRCNNTransform

Added more details to default __repr__ attribute for printing.

* fix: Put back relative imports

* style: Fixed pep8 compliance

Switched strings with  syntax to f-strings.

* test: Added test for GeneralizedRCNNTransform __repr__

Checked integrity of __repr__ attribute

* test: Fixed unittest for __repr__

Fixed the formatted strings in the __repr__ integrity check for GeneralizedRCNNTransform

* fix: Fixed f-strings for earlier python versions

Switched back f-strings to .format syntax for Python3.5 compatibility.

* fix: Fixed multi-line string

Fixed multiple-line string syntax for compatibility

* fix: Fixed GeneralizedRCNNTransform unittest

Fixed formatting of min_size argument of the resizing part

This reverts commit 28b7f8ae.

* Force object annotiation to be an array

* Remove unecessary parentheses

* Change object check

* Remove check for list

* Add test coverage to xml parsing

* Tidy up whitespace

* torchscriptable functions for video io (#1653)

Summary:
Pull Request resolved: https://github.com/pytorch/vision/pull/1653



created new torchscriptable video io functions as part of the api: read_video_meta_data_from_memory and read_video_from_memory.

Updated the implementation of some of the internal functions to be torchscriptable.

Reviewed By: stephenyan1231

Differential Revision: D18720474

fbshipit-source-id: 4ee646b66afecd2dc338a71fd8f249f25a3263bc

* BugFix
Co-authored-by: Jon Guerin <54725679+jguerin-fb@users.noreply.github.com>