MindYOLO代码易用性问题记录

[CaitinZhao]

If this is your first time, please read our contributor guidelines: https://gitee.com/mindspore/mindspore/blob/master/CONTRIBUTING.md

Is your feature request related to a problem? Please describe. A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]

Describe the solution you'd like

MindYOLO目前的版本存在一些语法的规避，和框架的易用性问题，在这个issue记录一下

Describe alternatives you've considered A clear and concise description of any alternative solutions or features you've considered.

Additional context Add any other context or screenshots about the feature request here.

[CaitinZhao]

1. loss部分冗余代码太多，比如 xywh2xyxy， to_tensor，在多个模型的loss中重复定义，需要提取公共函数

[CaitinZhao]

loss中使用stop_gradient太多，很难找到截断梯度的作用范围，后面2.0会提供类似torch的with torch.no_grad()方式

[CaitinZhao]

代码中使用了大量的tuple操作，后续改成python常用的list操作

[CaitinZhao]

现有的架构对loss的个数，输入的个数限制比较大，后续版本升级后改成dict的方式

[xunfeng2zkj]

不兼容mindspore-1.8.0

[zhiqwang]

有计划支持 ONNX 导出吗？