* support infer for SOLOv2

* unify log_smooth_windows and log_iter

test=develop

* remove duplicate log_iter

* remain mobile yml, since they are soft-link

test=develop

* share train_batch_size from reader to model

* update floor_divide

* add makedirs for dumping config

test=develop

* add solov2 model

* fix train batch size

* add solov2_r101_vd_fpn_3x

* fix batch size and update modelzoo

* refactor code of solov2

* fix deploy/python

* clean useless log in DataLoader

* fix infer set_sample_list_generator

test=develop

* add tools/anchor_cluster.py

reuse RetinaNet pipeline for now

* add PPYOLO
Co-authored-by: Nlongxiang <longxiang@baidu.com>

add htc model 

* add py_func check in export_model

* refine error message

add fleet training

Add image for visualdl board.

* Update docs
* Update README.md

* fix with_background in export model

* fix coco_eval background

* optimize python path

* add yolov4

* add serving export model
* add paddle serving client in requirement.txt
* remove redundant code with codestyle check
* fix export model doc
* fix according to professional reviews
* move paddle serving model export tutorial to the last of doc

* update check config

* fix slim/xx

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

current API is in flux, wait for it to stabilize

* Please CI

still buggy