Closed m-a-sch closed 4 months ago
Could you explain what you mean by "module decomposition"?
Probably not the right term. I mean this of course, the mlx classes and abstractions are not the same as the torch one, but the whisper abstractions should remain the same.
I understand your question now. Some things (e.g. triton_ops) don't make sense for MLX. Other things (e.g. torch_whipser and load_models are there to facilitate the conversion). Otherwise it's mostly the same with some simplifications. We don't intend for it to be a 1:1 match in terms of API and code organization. It should produce the same results though given the same audio.
Does MLX instead of torch really require to move away from the openAI module decomposition? Why?