add deform conv(#52)

add deform conv(#52)

 add deform conv (#52)

# add dcn in fcos_head.py(#95)

* support face detection in python inference

* support face detection in python deploy

* [cherry-pick] Fix v0.2 docs (#258)

* fix save checkpoint in quantization (#257)

* fix save checkpoint in quantization
* fix details

* Fix softlink (#256)
Co-authored-by: NLiufang Sang <slf12thuss@163.com>

* [cherry-pick] add prune benchmark (#262)

* add prune model latency (#260) & fix run command for windows (#261)

* Cherry-pick update reader config (#269)

* Force `cudnn` backend for depthwise convs when fp16 is enabled (#271)

* fix weights path (#274)

* cherry-pick refine fruit config (#289)

* cherry-pick polish doc (#287)

* cherry-pick fix oidv5 link (#296)

cherry-pick fix oidv5 link

* [cherry-pick] polish slim model zoo (#313)

* polish slim README

* refine pedestrian_yolov3_darknet.yml and vehicle_yolov3_darknet.yml in contrib (#324)

* fix yolo configs (#332)

* cherry pick polish doc (#329)

* refine config (#336)

* [cherry pick] refine sensitive link (#334)

* refine sensitive link

* add YOLOv3Loss.batch_size comments (#348)

* add deploy with python

* update README

* support TRT and Executor

* update cpp_infer for paddle latest

* support fixed shape of model

* polish code

* supoort trt

* support TRT&Executor

* deploy/python : fix coding style problems

* update requirements.txt

* update deploy/python docs
Co-authored-by: NGuanghua Yu <742925032@qq.com>
Co-authored-by: NLiufang Sang <slf12thuss@163.com>
Co-authored-by: NKaipeng Deng <dengkaipeng@baidu.com>
Co-authored-by: Nwangguanzhong <jerrywgz@126.com>
Co-authored-by: NYang Zhang <yangzhang@live.com>
Co-authored-by: littletomatodonkey <dazhiningsibuqu@163.com>
Co-authored-by: NFlyingQianMM <245467267@qq.com>
Co-authored-by: NChanningss <chen_lingchi@163.com>

* 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 mask operations in RandomExpand and RandomCrop

* remove all [] in cropped polys

* is_poly -> _is_poly()

* change the judgment on whether there are polygons

* abstract is_poly to op_helper.py

* add is_mask_expand and is_mask_crop in RandomExpand and RamdomCrop

* add mask operations in RandomExpand and RandomCrop

* remove all [] in cropped polys

* is_poly -> _is_poly()

* change the judgment on whether there are polygons

* abstract is_poly to op_helper.py

* 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%

* update cpp_infer for paddle latest

* polish code

* export config file in export_model.py

* add custom label_list & dump_box

* add infer_cfg.yml & doc

* export label_list to inference config

* update PaddleTensor to ZeroCopyTensor

Add cosine decay lr strategy.