The GPU device does not support Double (Float64) operations

E:\RVC0813AMD_Intel>runtime\python.exe infer-web.py --pycmd runtime\python.exe --port 7897 --dml
No supported Nvidia GPU found
use DirectML instead
Use Language: zh_CN
Running on local URL: http://0.0.0.0:7897
loading weights/guanguanV1.pth
gin_channels: 256 self.spk_embed_dim: 109

loading rmvpe model
2023-08-31 23:06:29.8257221 [W:onnxruntime:, session_state.cc:1169 onnxruntime::VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2023-08-31 23:06:29.8346421 [W:onnxruntime:, session_state.cc:1171 onnxruntime::VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
cleaning ortruntime memory
Traceback (most recent call last):
File "E:\RVC0813AMD_Intel\infer-web.py", line 211, in vc_single
audio_opt = vc.pipeline(
File "E:\RVC0813AMD_Intel\vc_infer_pipeline.py", line 359, in pipeline
pitchf = torch.tensor(pitchf, device=self.device).unsqueeze(0).float()
RuntimeError: The GPU device does not support Double (Float64) operations!

Traceback (most recent call last):
File "E:\RVC0813AMD_Intel\runtime\lib\site-packages\gradio\routes.py", line 321, in run_predict
output = await app.blocks.process_api(
File "E:\RVC0813AMD_Intel\runtime\lib\site-packages\gradio\blocks.py", line 1007, in process_api
data = self.postprocess_data(fn_index, result["prediction"], state)
File "E:\RVC0813AMD_Intel\runtime\lib\site-packages\gradio\blocks.py", line 953, in postprocess_data
prediction_value = block.postprocess(prediction_value)
File "E:\RVC0813AMD_Intel\runtime\lib\site-packages\gradio\components.py", line 2076, in postprocess
processing_utils.audio_to_file(sample_rate, data, file.name)
File "E:\RVC0813AMD_Intel\runtime\lib\site-packages\gradio\processing_utils.py", line 206, in audio_to_file
data = convert_to_16_bit_wav(data)
File "E:\RVC0813AMD_Intel\runtime\lib\site-packages\gradio\processing_utils.py", line 219, in convert_to_16_bit_wav
if data.dtype in [np.float64, np.float32, np.float16]:
AttributeError: 'NoneType' object has no attribute 'dtype'

这是显卡不支持还是模型的问题呀😆

Start with vc_infer_pipeline.py at line 359 and trace the pitchf value passed into torch.tensor, then review infer-web.py around vc_single. Reproduce the failure with the shown DirectML command and verify that voice conversion completes without the Float64 error or the subsequent NoneType error.