* add code comments
* merge *_saver.py & *_loader.py in io/
* (ancient codes) rename Loss into LossFromTorch

* clean up unused codes

* 更新Loss的接口形参跟metric保持一致
* 添加对几种loss的测试
* embed_loader采用维度独立的方法采样
* 对应测试代码的修改

* improve Loss initialization interface
* improve test codes for trainer

* fix bug in DataSet.split
* fix bugs in FieldArray, to allow content as a list
* fix bug in losses check
* ...