* add python train time eval

* add mpii infer support

Besides removing unnecessary imports, this also fixes the missing logger module in test_architectures.py within the static directory.

* merge upstream

* clean code

* add batch_size in keypoint_detector

* support --eval for s2anet

* add spine dataset and config yml for s2anet

* add doc for s2anet

* update doc

* fix typo, test=document_fix

* add some comments, and update import

* updatee pretrain match

* update match for rcnn

* update version and config download

* add config for jde and deepsort, add engine metric tools

* add mot_metrics

* fix configs engine and others

* fix configs others

* move configs

* fix ap_per_class, fix doc

* fix test_mode, seqs

* fix test_mode to metric

* fix JDE arch metric

* clean requirement

* add to requirement

* Add BottomUp HigherHRNet for pose.
* reverse xtcocotools to pycocotools

* update 2.0.1

* fix some docs error

* fix name of pretrain weights

* format, test=document_fix

* Add S2ANet model for Oriented Object Detection.

* refine ppdet file name

* unify bbox_utils

* add Distill model
* add distill+prune

* support mask_rcnn for kunlun

* minor

* update solov2 some detail

* fix check paddle version

* check md5 before load weights.

* support infer for SOLOv2

* 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

* check md5 after weights download

* fix voc mAP %.