fix(demo): typo in README

fix(wandb): key messed in wandb option parsing

docs(README): use absolute URLs for model names

chore(README): add StreamYOLO.

tag(YOLOX): bump to 0.3

fix(README & model) some typos

fix(README):  typo

fix(demo/TensorRT): except in logging.h

fix(README): link to `yolox_base.py`

fix(README):  correct onnx url path of yolox_x

feat(model): support hub load

chore(REAdME): add HuggingFace web demo link

* feat(utils): freeze module

fix(exps):  typo of base_exp

feat(tools): add option for decode output in export_onnx

feat(engine): support wandb logger

fix(data): dataset error when running on test-dev

fix(utils): occupy_mem based on CUDA_VISIBLE_DEVICES

chore(tools): show demo in live window

chore(evaluator): support  custom dataset
Co-authored-by: NFeng Wang <wangfeng19950315@163.com>

fix(exps): typo in tiny

fix(data): value convert error

fix(demo & tools): path issues of Windows system

chore(data): Adapt to COCO format dataset

* feat(evaluator): support logging per class mAP during evaluation

* make linter happy

This reverts commit 2c2dd139.

chore(setup): add torch to build requirements