Open SetoKaiba opened 1 month ago
see 1241
是不是合并了就解决啦?
Yes
@SetoKaiba 当出现并行请求时,仍然会出现这个问题,会被自动更改为CPU推理,不知道为什么。
@SetoKaiba 当出现并行请求时,仍然会出现这个问题,会被自动更改为CPU推理,不知道为什么。
How to reproduce?
@SetoKaiba 当出现并行请求时,仍然会出现这个问题,会被自动更改为CPU推理,不知道为什么。
How to reproduce?
@SetoKaiba
https://github.com/RVC-Boss/GPT-SoVITS/assets/47490867/89fa3985-c934-4590-9f10-31d97ad497b6
Sorry for delayed response, I just simply send two requests in one go, and the server used cpu infer as fallback.
It seems that the move_to_gpu & move_to_cpu is not working as expected in the branch fast_inference.
https://github.com/RVC-Boss/GPT-SoVITS/blob/fast_inference_/api_v3.py#L327-L343
It will always move to cpu. The custom settings device is cuda. The precision is half.
When the move_to_cpu is called, the custom settings device became cpu and the precision kept half. But according to lines below, half is not supported for cpu.
https://github.com/RVC-Boss/GPT-SoVITS/blob/fast_inference_/GPT_SoVITS/TTS_infer_pack/TTS.py#L311-L313
I try to add the lines below to move_to_cpu. https://github.com/RVC-Boss/GPT-SoVITS/blob/fast_inference_/api_v3.py#L348-L351
I try to add the lines below to move_to_gpu. https://github.com/RVC-Boss/GPT-SoVITS/blob/fast_inference_/api_v3.py#L354-L356
But then the first tts request is working. The second isn't. Here's the error of the second request.
@KevinZhang19870314