You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have been able to train and infer on this computer previously but I am running out of memory inferring a particular wav file (I have successfully processed longer wav files previously). Are the duration and sample rate of the wav file affecting memory usage?
I am using default parameters.
C:\Mangio-RVC-v23.7.0>runtime\python.exe infer-web.py --pycmd runtime\python.exe --port 7897
Found GPU NVIDIA GeForce RTX 2060
Set fp16_run to true in 32k.json
Set fp16_run to true in 40k.json
Set fp16_run to true in 48k.json
Use Language: en_US
Running on local URL: http://0.0.0.0:7897
To create a public link, set share=True in launch().
loading weights/Benny.pth
gin_channels: 256 self.spk_embed_dim: 109
loading rmvpe model
Traceback (most recent call last):
File "C:\Mangio-RVC-v23.7.0\infer-web.py", line 316, in vc_single
audio_opt = vc.pipeline(
File "C:\Mangio-RVC-v23.7.0\vc_infer_pipeline.py", line 564, in pipeline
self.vc(
File "C:\Mangio-RVC-v23.7.0\vc_infer_pipeline.py", line 450, in vc
(net_g.infer(feats, p_len, pitch, pitchf, sid)[0][0, 0])
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\models.py", line 752, in infer
m_p, logs_p, x_mask = self.enc_p(phone, pitch, phone_lengths)
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\models.py", line 104, in forward
x = self.encoder(x * x_mask, x_mask)
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\attentions.py", line 65, in forward
y = self.attn_layers[i](x, x, attn_mask)
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\attentions.py", line 221, in forward
x, self.attn = self.attention(q, k, v, mask=attn_mask)
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\attentions.py", line 265, in attention
relative_weights = self._absolute_position_to_relative_position(p_attn)
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\attentions.py", line 346, in _absolute_position_to_relative_position
x_flat = F.pad(x_flat, commons.convert_pad_shape([[0, 0], [0, 0], [length, 0]]))
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 528.00 MiB (GPU 0; 6.00 GiB total capacity; 3.75 GiB already allocated; 0 bytes free; 4.81 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
Traceback (most recent call last):
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\routes.py", line 437, in run_predict
output = await app.get_blocks().process_api(
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\blocks.py", line 1349, in process_api
data = self.postprocess_data(fn_index, result["prediction"], state)
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\blocks.py", line 1283, in postprocess_data
prediction_value = block.postprocess(prediction_value)
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\components.py", line 2586, in postprocess
file_path = self.audio_to_temp_file(
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\components.py", line 360, in audio_to_temp_file
temp_dir = Path(dir) / self.hash_bytes(data.tobytes())
The text was updated successfully, but these errors were encountered:
After reading some of the other bug reports here I was able to get it working by setting the following parameters in config.py: x_pad = 1 x_query = 5 x_center = 30 x_max = 32
Are these parameters explained anywhere?
The calculation of GPU memory confused me a little, why is 0.4 added to self.gpu_mem?
Maybe it would be better to subtract something here instead to take into account gpu memory used by other programs and have a safety margin?
I have been able to train and infer on this computer previously but I am running out of memory inferring a particular wav file (I have successfully processed longer wav files previously). Are the duration and sample rate of the wav file affecting memory usage?
I am using default parameters.
C:\Mangio-RVC-v23.7.0>runtime\python.exe infer-web.py --pycmd runtime\python.exe --port 7897
Found GPU NVIDIA GeForce RTX 2060
Set fp16_run to true in 32k.json
Set fp16_run to true in 40k.json
Set fp16_run to true in 48k.json
Use Language: en_US
Running on local URL: http://0.0.0.0:7897
To create a public link, set
share=True
inlaunch()
.loading weights/Benny.pth
gin_channels: 256 self.spk_embed_dim: 109
loading rmvpe model
Traceback (most recent call last):
File "C:\Mangio-RVC-v23.7.0\infer-web.py", line 316, in vc_single
audio_opt = vc.pipeline(
File "C:\Mangio-RVC-v23.7.0\vc_infer_pipeline.py", line 564, in pipeline
self.vc(
File "C:\Mangio-RVC-v23.7.0\vc_infer_pipeline.py", line 450, in vc
(net_g.infer(feats, p_len, pitch, pitchf, sid)[0][0, 0])
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\models.py", line 752, in infer
m_p, logs_p, x_mask = self.enc_p(phone, pitch, phone_lengths)
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\models.py", line 104, in forward
x = self.encoder(x * x_mask, x_mask)
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\attentions.py", line 65, in forward
y = self.attn_layers[i](x, x, attn_mask)
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\attentions.py", line 221, in forward
x, self.attn = self.attention(q, k, v, mask=attn_mask)
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\attentions.py", line 265, in attention
relative_weights = self._absolute_position_to_relative_position(p_attn)
File "C:\Mangio-RVC-v23.7.0\lib\infer_pack\attentions.py", line 346, in _absolute_position_to_relative_position
x_flat = F.pad(x_flat, commons.convert_pad_shape([[0, 0], [0, 0], [length, 0]]))
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 528.00 MiB (GPU 0; 6.00 GiB total capacity; 3.75 GiB already allocated; 0 bytes free; 4.81 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
Traceback (most recent call last):
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\routes.py", line 437, in run_predict
output = await app.get_blocks().process_api(
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\blocks.py", line 1349, in process_api
data = self.postprocess_data(fn_index, result["prediction"], state)
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\blocks.py", line 1283, in postprocess_data
prediction_value = block.postprocess(prediction_value)
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\components.py", line 2586, in postprocess
file_path = self.audio_to_temp_file(
File "C:\Mangio-RVC-v23.7.0\runtime\lib\site-packages\gradio\components.py", line 360, in audio_to_temp_file
temp_dir = Path(dir) / self.hash_bytes(data.tobytes())
The text was updated successfully, but these errors were encountered: