async algorithm conflict with fused optimizer, can only support flattened inplac...

[Al]

bagua/bagua/torch_api/algorithms/async_model_average.py

Lines 84 to 89 in 8707188

 # TODO: async algorithm conflict with fused optimizer, can only support flattened inplace bucket. 
 assert do_flatten, "async does not support `do_flatten=False` at present." 
 if self.step_id < self.warmup_steps: 
     return super().tensors_to_buckets(tensors, do_flatten) 
  
 all_tensors = [] 

---

This issue was generated by todo based on a TODO comment in 8707188. It's been assigned to @wangraying because they committed the code.