cast dtype in load_pretrain_weight

* add yolov3 PrunerQAT joint compression strategy

* update readme

* Implementation of WandbLogger callback to log metrics and model checkpoints

* Added documentation on how to use WandbLogger and VisualDL

* Minor bug fix and pull from upstream

* Removed pass statement for cleanup

* Update logging_en.md

Fixed documentation

* Update trainer.py

* polish args

* fix import bugs

* fix yaml constructor in eval & export

* fix typo

* save infer results to file.

* update, for mac temp file

* update func name, fix mem, fix dir

* add yolox doc and configs

* fix bn and update doc, test=document_fix

* revert det tqdm eval, test=document_fix

* revert det tqdm eval, test=document_fix

* support nms options when export model

* fix config

* add bytetrack yolov3 engine and deploy baseline

* fix bytetrack yolov3 configs

* fix cfgs tools

* fix tracker pred_dets

* fit pptracking jde sde infer

* rename cfgs, fix mot_sde_infer

* fix mot sde single class infer

* remove byte cfgs

* fix mtmct

* clean code

* add mcfiarmot ptq

* update docs

* [ppyolov3] add ppyolov3 base code

* add ppyolov3 s/m/x

* modify ema

* modify code to convert onnx successfully

* support arbitrary shape

* update config to use amp default

* refine ppyolo_head code

* modify reparameter code

* refine act layer

* adapter pico_head and tood_head code

* remove ppyolov3 yaml

* fix codestyle
Co-authored-by: Nwangxinxin08 <wangxinxin08@baidu.com>

This reverts commit 3e0e3585.

* support export post_process in PicoDet

* support different norm_type for different module when convert to sync batch norm

* remove BatchNorm in esnet

* refine sync bn

* fix bugs of batch_norm

* fix bugs while deploying and modify BatchNorm to BatchNorm2D

* param_attr -> weight_attr in BatchNorm2D

* modify BatchNorm to BatchNorm2D