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

* add multi-scale testing function in cascade rcnn clsaware architecture

* exit1 in travis failed

* Add GIOU loss.
* Add DIOU/CIOU loss.
* Change loss box type in bbox_head.py.
* Remove fuse_elewise_add_act_ops flag in train.py (Error may happen for elem_divbased on the flag).

* add softnms, nonlocal, resnet200_vd_backbone
* add CBNet
* update model zoo

* add multi-scale test

* add cascade gn

* add cascade mask gn config

* update configs

* refine cascade models

* add a global norm_type flag

* add group norm on fpn

* add gn on box head

* add faster rcnn fpn 50 gn config

* add mask branch norm

* update configs

* Add full configs example & refine doc

* refine docs

* Clean up a bit

* Add `__shared__` annotation for shared variables between modules

* Dedup `num_classes` configurations

* Document `__shared__` usage

* Remove some unused variable

* Improve docstring and add comments for `__shared__`

* Rename object_detection to PaddleDetection
* Small fix for doc