* update mbv3 ssdlite

* fix ssdlite link

Add practical generic detection models containing 676 categories

* update speed

* update model size of blazeface nas_v2

* remove speed of retinanet

* update doc

* Add cutmix op(#88)

* split weights and PaddleLite model in configs/mobile/README

* Add ACfpn, `CxAM` and `CnAM` is not included here.

* add version check

* update version check & python version for cornerpool

* activate scale_x_y

* update cspdarknet pretrained model

* Add libra-rcnn, GCNet and iou loss intro doc.

* fix export_model doc link error

* sort out release/0.3 docs

* add new CHANGELOG

* add yolov4 in changelog

* add detailed doc
* add zh_CN doc of pssdet
* fix link

* add yolov4

* Organize and document mobile models

* Clean up trailing whitespaces

* Change bullet type

* Minor wording fixes

* Fix yolo naming

* Add link in repo README.md

use_fine_grained_loss: true

* Revert back to `fluid.clip` for gradient clipping

current API is in flux, wait for it to stabilize

* Please CI

still buggy

* add config link in model zoo

* fix link error

* add anchor free

* #add dcn on  FCOS_head and backbone

* code for cornernet

* add doc

* refine custom op

* refine code from review comments

* refine doc

* update doc

* update code

* add cornernet_squeeze_mixup_cosine config

* update code

* update code

* add moile-side rcnn server model

* add iou aware module
* iouaware model zoo
* fix copyright

* add mobilenet v3 configs

* fix some comment

* fix memsize

* fcos initialization

* delete the codes not used

* pre-commit the committed code

* delete the unused function

* modify the capacity in loader from 64 to 16

* add devices_num on reader

* refine capacity value

* refine bufsize

fix rcnn server side rcnn models' pretrain path

* add yolov3_mobilenet_v3

* add server side rcnn model, 62 fps with coco mAP 41.6% and 20fps with coco mAP 47.8%