* Add fsaf into regression benchmark

* Cover more cases in unittest & regression benchmark

* Minor fix

* rm nas_fpn from test_pipeline

* avoid duplicated cfgs

* add configs for all models

* fix some configs

* Add benchmark models

* update some benchmarks

* Fix retina config

* add more benchmarks

* update MODEL_ZOO.md

* fix cascade mask num_class bug

* change caffe pretrained url

* fix paramwise_cfg bug

* fix paramwise cfg

* restrict minimum h and w in crop_and_resize

* add grad clip for fcos

* add grad clip for ga retinanet

* fix grad clip

* fix img norm

* fix rpn loss bbox

* fix rpn proposal num

* fix url

* clean

* update comparison

* minor update

* add grad clip for guided anchoring

* fix some typos

* add inference time details

* clean

Co-authored-by: ZwwWayne <wayne.zw@outlook.com>

* Add tutorials for V2.0

* Add details for dataset tutorial

* add tutorials in index.rst

* Add tutorials for concat data

* Modify documents

* More contents

* starting to merge fsaf

* changed configs

* changed bbox_coder

* get_targets fun in AnchorHead class is modified to enable additional returns from _get_target_single

* effective area assigner and tblr bbox_coder are implemented

* added fsaf_head and config

* added more fsaf configs

* fixed the nan bbox_loss by clamping bbox_pred with min=1e-4

* fixed the nan bbox_loss by clamping bbox_pred with min=1e-4

* changed config

* formatter

* changed docstring

* Added ignored pixel-class pair for FSAF

* More elegant assignment for overlapping objects

* more documents in effective area assigner

* fixed a bug in object overlay

* Added no gt_bboxes case and pytest for assigner

* Added no gt_bboxes case and pytest for assigner

* changed default config

* fixed small bugs in FSAF head doc example

* resolved pr review comments.
1. Formats of comments and docstrings are modified
2. Obsolete arguments are removed from FSAFHead
3. AssignResults class is revised to be more compatible with old version

* Added more comments to FSAFHead

* resolved pr review comments

* resolved pr review comments
1. More flexible CenterRegionAssigner bbox priority
2. More flexible tblr_coder by adding normalize_by_wh flag
3. README modifications with model link
4. More docstrings and comments
5. Reformatted
6. Change of default normalizer in TBLRBBoxCoder from 1.0 to 4.0

* Corrected FSAF model path in readme

* Fixed the bug of bbox weights in fsaf head

Co-authored-by: wangxinjiang <wangxinjiang@sensetime.com>

* Pass momentum_config to register_training_hooks

* Bump mmcv version

* Explicit default

V2.0 Merge Master

Conflicts:
	mmdet/datasets/xml_style.py

* add model upgrade tool

* Add configs and scripts for converting v1.x models to v2.0

* add doc

* Add more models

* Pass CI

* Clean scripts and check model by config

* Resolve comments

* Resolve small comments

Co-authored-by: Jiarui XU <xvjiarui0826@gmail.com>

* fixed iou_loss

* add weight=None check

* add head-iou

* fixed comment

* add shape

* change view to asset

* Use apis/inference in detectors show_result

* Add images_out_dir arg to test apis

* Include --images_out_dir in assertion

* Fix single class parsing

* Remove single class hack

* Move import

* Add example of saving results

* Fix list number

* Refactor show_result

* Update docs

* Update __init__

* Fix CLASSES reference

* Remove unnecessar assert for custom classes

* Use - instead of _

* Use - instead of _

* Rename images_out_dir -> out_dir

* Remove unnecessary model.module

* Update show_result with all params and docstring

* Fix missing comma

* Set rescale always to True

* Remove outdated restriction from docs

* Drop pathlib

* Rename out_dir -> show_dir

* Update docstring

* More explicit code. Fix out_dir

* Update docstrings

* Flake8

Co-authored-by: mmeendez8 <miguelmndez@gmail.com>
Co-authored-by: sbugallo <sbugallo@gradiant.org>

Co-authored-by: z-bingo <z-bingo@outlook.com>

* fix unmap

* make it more clear

* rm bboxes[:,:4]

* Move the bbox size selection and check into iou2d_calculator

* Add bbox shape check in bbox_overlaps

* Validate the bbox sizes

* add docs for new config

* update text->pretty_text, replace MultiKVAction with DictAction

* update missing

* update docs

* added background label to assign function

* fixed the background label as -1 in all assigners

Co-authored-by: wangxinjiang <wangxinjiang@sensetime.com>

* consummating xml_style.py

* Update xml_style.py

update xml_style.py

* Update xml_style.py

fix the format

* Fix bugs in regression test

* Fix test_loading

* Add scripts to filter out regression benchmark configs

* Use 1x rather than 2x schedule

* delete duplicated script

* Fix typo

* Change .format to f-string

* Resolve comments

* Fix missing change

* replace AT_CHECK with TORCH_CHECK to support pytorch1.5

* fix compile warning, delete grid sample op

* fix ci pytorch version

* delete python3.5

* update version of pytorch and torchvision in ci

* change is_cuda to device().is_cuda()

* specify TORCH_CUDA_ARCH_LIST=6.0,7.0

* fix typo

* delete affine grid

* fix typos

* fix setup.py

* remove redundant comment

* Add documentation

* Fix (doc): fix syntax error

* Add details about hyper parameters

* Add name convention

* Update compatibility docs

* add dilation of conv_offset

* Update deform_conv.py

* use mmcv Registry instead of mmdet

* merged registry and builder

* merge dataset builder registry

* remove circulate import in bbox

* rename registry to builder

Merge master into V2.0

* Support V1.0 anchor generator

* Reformat code

* Add registry for anchor generator

* Reformat docstring

* fix isort

* Refactor anchor generator

* reformat doc, add unittest

* Refactor retinahead anchor generator

* Fix test_anchor device bug

* Fix docstring

* Resolve comments with more unittests

* Fix atss bug and add more unittest

* Resolve comments

* Add more asset and check

* fixed init/refine

* fixed bbox coder assert

* add reg_decoded_bbox in two_stage

* finished single stage

* fixed comment

* fixed docs