GitLab

* intial structure

* fixed types of few variables

* remove the commented code

* list -> List

* encode method will take input as tensors instead of list of tensor
Co-authored-by: Joao Gomes <jdsgomes@fb.com>

* upgrade usort to

* Also update black

* Actually use 1.0.2

* Apply pre-commit
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* basic structure

* added constrains

* fixed errors

* thanks to vadim!

* addressing the comments and added docstrign

* Apply suggestions from code review
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* Adding ciou and diou support in `_box_loss()`

* Fix linter

* Addressing comments for nits

* Use frozen BN only if pre-trained.

* Add LSJ and ability to from scratch training.

* Fixing formatter

* Adding `--opt` and `--norm-weight-decay` support in Detection.

* Fix error message

* Make ScaleJitter proportional.

* Adding more norm layers in split_normalization_params.

* Add FixedSizeCrop

* Temporary fix for fill values on PIL

* Fix the bug on fill.

* Add RandomShortestSize.

* Skip resize when an augmentation method is used.

* multiscale in [480, 800]

* Add missing star

* Add new RetinaNet variant.

* Add tests.

* Update expected file for old retina

* Fixing tests

* Add FrozenBN to retinav2

* Fix network initialization issues

* Adding BN support in MaskRCNNHeads and FPN

* Adding support of FasterRCNNHeads

* Introduce norm_layers in backbone utils.

* Bigger RPN head + 2x rcnn v2 models.

* Adding gIoU support to retinanet

* Fix assert

* Add back nesterov momentum

* Rename and extend `FastRCNNConvFCHead` to support arbitrary FCs

* Fix linter

* reverting some recently introduced exceptions

* Update torchvision/ops/poolers.py
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* address PR comments

* replace one more assert with torch._assert:

* address PR comments

* make type checker happy

* Fix bug

* fix bug

* fix for wrong asserts

* attempt to make tests pass

* Fix test_ops tests

* Fix expected exception in tests

* fix typo

* fix tests and format

* fix flake8

* remove one last exception

* fix error

* remove unused immport

* replace fake returns by else
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* replace most asserts with exceptions

* fix formating issues

* fix linting and remove more asserts

* fix regresion

* fix regresion

* fix bug

* apply ufmt

* apply ufmt

* fix tests

* fix format

* fix None check

* fix detection models tests

* non scriptable any

* add more checks for None values

* fix retinanet test

* fix retinanet test

* Update references/classification/transforms.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Update references/classification/transforms.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Update references/optical_flow/transforms.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Update references/optical_flow/transforms.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Update references/optical_flow/transforms.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* make value checks more pythonic:

* Update references/optical_flow/transforms.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* make value checks more pythonic

* make more checks pythonic

* fix bug

* appy ufmt

* fix tracing issues

* fib typos

* fix lint

* remove unecessary f-strings

* fix bug

* Update torchvision/datasets/mnist.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Update torchvision/datasets/mnist.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Update torchvision/ops/boxes.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Update torchvision/ops/poolers.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Update torchvision/utils.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* address PR comments

* Update torchvision/io/_video_opt.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Update torchvision/models/detection/generalized_rcnn.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Update torchvision/models/feature_extraction.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Update torchvision/models/optical_flow/raft.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* address PR comments

* addressing further pr comments

* fix bug

* remove unecessary else

* apply ufmt

* last pr comment

* replace RuntimeErrors
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* Use onnx function only in tracing mode

* Add missing import

* Address review comments

* Fix type annotation

* Ignore return type error

* Remove unused import

* Add fake cast

* Fix mypy error

* Fix mypy error

* Update torchvision/models/detection/_utils.py
Co-authored-by: Shubham Bhokare <32080845+shubhambhokare1@users.noreply.github.com>

* _fake_cast_onnx approach
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>
Co-authored-by: Shubham Bhokare <32080845+shubhambhokare1@users.noreply.github.com>

* Add topk minimizer function to _utils

* Apply ufmt formatting

* Apply min function for tracing and scripting

* Add type ignore to avoid cast

* fix flake

* Fix python_type_check
Co-authored-by: Prabhat Roy <prabhatroy@fb.com>

* add fcos

* update fcos

* add giou_loss

* add BoxLinearCoder for FCOS

* add full code for FCOS

* add giou loss

* add fcos

* add __all__

* Fixing lint

* Fixing lint in giou_loss.py

* Add typing annotation to fcos

* Add trained checkpoints

* Use partial to replace lambda

* Minor fixes to docstrings

* Apply ufmt format

* Fixing docstrings

* Fixing jit scripting

* Minor fixes to docstrings

* Fixing jit scripting

* Ignore mypy in fcos

* Fixing trained checkpoints

* Fixing unit-test of jit script

* Fixing docstrings

* Add test/expect/ModelTester.test_fcos_resnet50_fpn_expect.pkl

* Fixing test_detection_model_trainable_backbone_layers

* Update test_fcos_resnet50_fpn_expect.pkl

* rename stride to box size

* remove TODO and fix some typo

* merge some code for better

* impove the comments

* remove decode and encode of BoxLinearCoder

* remove some unnecessary hints

* use the default value in detectron2.

* update doc

* Add unittest for BoxLinearCoder

* Add types in FCOS

* Add docstring for BoxLinearCoder

* Minor fix for the docstring

* update doc

* Update fcos_resnet50_fpn_coco pretained weights url

* Update torchvision/models/detection/fcos.py
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* Update torchvision/models/detection/fcos.py
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* Update torchvision/models/detection/fcos.py
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* Update torchvision/models/detection/fcos.py
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* Add FCOS model documentation

* Fix typo in FCOS documentation

* Add fcos to the prototype builder

* Capitalize COCO_V1

* Fix params of fcos

* fix bug for partial

* Fixing docs indentation

* Fixing docs format in giou_loss

* Adopt Reference for GIoU Loss

* Rename giou_loss to generalized_box_iou_loss

* remove overwrite_eps

* Update AP test values

* Minor fixes for the docs

* Minor fixes for the docs

* Update torchvision/models/detection/fcos.py
Co-authored-by: Zhiqiang Wang <zhiqwang@foxmail.com>

* Update torchvision/prototype/models/detection/fcos.py
Co-authored-by: Zhiqiang Wang <zhiqwang@foxmail.com>
Co-authored-by: zhiqiang <zhiqwang@foxmail.com>
Co-authored-by: Joao Gomes <jdsgomes@fb.com>
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>
Co-authored-by: Joao Gomes <joaopsgomes@gmail.com>

Co-authored-by: Nicolas Hug <nicolashug@fb.com>

* Start annotating utils

* checking

* Add annotations at _utils.py

* Remove unnecessary comments.

* re-checked typings

* Update typing

* Ignore small error

* Use optional tensor

* Ignore for JIT
Co-authored-by: Khushi Agrawal <khushiagrawal411@gmail.com>
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* add ufmt as code formatter

* cleanup

* quote ufmt requirement

* split imports into more groups

* regenerate circleci config

* fix CI

* clarify local testing utils section

* use ufmt pre-commit hook

* split relative imports into local category

* Revert "split relative imports into local category"

This reverts commit f2e224cde2008c56c9347c1f69746d39065cdd51.

* pin black and usort dependencies

* fix local test utils detection

* fix ufmt rev

* add reference utils to local category

* fix usort config

* remove custom categories sorting

* Run pre-commit without fixing flake8

* got a double import in merge
Co-authored-by: Nicolas Hug <nicolashug@fb.com>

* Explicitely store a distance value that is reused 

I don't see a reason to calculate the value twice for each distance. 

Knowing that this code is going to be called at every epoch and that probably there is not a compiler that optimizes this (+ also code clarity), I think is best to store the value for x and y.

* Update torchvision/models/detection/_utils.py
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* removing spaces
Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* Early skeleton of API.

* Adding MultiFeatureMap and vgg16 backbone.

* Making vgg16 backbone same as paper.

* Making code generic to support all vggs.

* Moving vgg's extra layers a separate class + L2 scaling.

* Adding header vgg layers.

* Fix maxpool patching.

* Refactoring code to allow for support of different backbones & sizes:
- Skeleton for Default Boxes generator class
- Dynamic estimation of configuration when possible
- Addition of types

* Complete the implementation of DefaultBox generator.

* Replace randn with empty.

* Minor refactoring

* Making clamping between 0 and 1 optional.

* Change xywh to xyxy encoding.

* Adding parameters and reusing objects in constructor.

* Temporarily inherit from Retina to avoid dup code.

* Implement forward methods + temp workarounds to inherit from retina.

* Inherit more methods from retinanet.

* Fix type error.

* Add Regression loss.

* Fixing JIT issues.

* Change JIT workaround to minimize new code.

* Fixing initialization bug.

* Add classification loss.

* Update todos.

* Add weight loading support.

* Support SSD512.

* Change kernel_size to get output size 1x1

* Add xavier init and refactoring.

* Adding unit-tests and fixing JIT issues.

* Add a test for dbox generator.

* Remove unnecessary import.

* Workaround on GeneralizedRCNNTransform to support fixed size input.

* Remove unnecessary random calls from the test.

* Remove more rand calls from the test.

* change mapping and handling of empty labels

* Fix JIT warnings.

* Speed up loss.

* Convert 0-1 dboxes to original size.

* Fix warning.

* Fix tests.

* Update comments.

* Fixing minor bugs.

* Introduce a custom DBoxMatcher.

* Minor refactoring

* Move extra layer definition inside feature extractor.

* handle no bias on init.

* Remove fixed image size limitation

* Change initialization values for bias of classification head.

* Refactoring and update test file.

* Adding ResNet backbone.

* Minor refactoring.

* Remove inheritance of retina and general refactoring.

* SSD should fix the input size.

* Fixing messages and comments.

* Silently ignoring exception if test-only.

* Update comments.

* Update regression loss.

* Restore Xavier init everywhere, update the negative sampling method, change the clipping approach.

* Fixing tests.

* Refactor to move the losses from the Head to the SSD.

* Removing resnet50 ssd version.

* Adding support for best performing backbone and its config.

* Refactor and clean up the API.

* Fix lint

* Update todos and comments.

* Adding RandomHorizontalFlip and RandomIoUCrop transforms.

* Adding necessary checks to our tranforms.

* Adding RandomZoomOut.

* Adding RandomPhotometricDistort.

* Moving Detection transforms to references.

* Update presets

* fix lint

* leave compose and object

* Adding scaling for completeness.

* Adding params in the repr

* Remove unnecessary import.

* minor refactoring

* Remove unnecessary call.

* Give better names to DBox* classes

* Port num_anchors estimation in generator

* Remove rescaling and fix presets

* Add the ability to pass a custom head and refactoring.

* fix lint

* Fix unit-test

* Update todos.

* Change mean values.

* Change the default parameter of SSD to train the full VGG16 and remove the catch of exception for eval only.

* Adding documentation

* Adding weights and updating readmes.

* Update the model weights with a more performing model.

* Adding doc for head.

* Restore import.

* Document undodcumented parameters

* remove setup.cfg changes

* Properly pass normalize down instead of deprecating it

* Fix flake8

* Add new CI check

* Fix type spec

* Leave normalize be part of kwargs
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* Introduce small score threshold on rpn

* Adding docs and fixing keypoint and mask.

* Making value 0.0 by default for BC.

* Fixing for onnx.

* Update threshold.

* Removing non-default threshold from reference scripts.
Co-authored-by: Francisco Massa <fvsmassa@gmail.com>

Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* Replacing all torch.jit.annotations with typing

* Replacing remaining typing

* Overwriting FrozenBN eps=0.0 if pretrained=True for detection models.

* Moving the method to detection utils and adding comments.

Replace nonzero by where, now that it works with just a condition

* Try remove eager scripting calls

* remove script call
Co-authored-by: eellison <eellison@fb.com>
Co-authored-by: Francisco Massa <fvsmassa@gmail.com>

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

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

* Fix mypy type annotations

* follow torchscript Tuple type

* redefine torch_choice output type

* change the type in cached_grid_anchors

* minor bug
Co-authored-by: Guanheng Zhang <zhangguanheng@devfair0197.h2.fair>
Co-authored-by: Guanheng Zhang <zhangguanheng@learnfair0341.h2.fair>

* For detection model, replace L1 loss with beta smooth L1 loss to achieve better performance.

* Add type annotations for torchscript

* Resolve E226

* remove sys.version_info == 2

* remove sys.version_info < 3

* remove from __future__ imports

* almost working...

* respond to comments

* add empty tensor op, handle different output types in generalized rcnn

* clean ups

* address comments

* more changes

* it's working!

* torchscript bugs

* add script/ eager test

* eval script model

* fix flake

* division import

* py2 compat

* update test, fix arange bug

* import division statement

* fix linter

* fixes

* changes needed for JIT master

* cleanups

* remove imagelist_to

* requested changes

* Make FPN backwards-compatible and torchscript compatible

We remove support for feature channels=0, but support for it was already a bit limited

* Fix ONNX regression

* Support Exporting RPN to ONNX

* address PR comments

* fix cat

* add flatten

* replace cat by stack

* update test to run only on rpn module

* use tolerate_small_mismatch

* [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