Closed 729533572 closed 1 year ago
windows 11
4090
3.10
n/a
4.0(默认)
个人声音处理
推理
训练2400步后,生成的模型,在预处理的时候报错 异常信息:Given groups=1, weight of size [192, 768, 5], expected input[1, 256, 773] to have 768 channels, but got 256 channels instead 请排障后重试
Traceback (most recent call last): File "app.py", line 150, in vc_fn _audio = model.slice_inference(temp_path, sid, vc_transform, slice_db, cluster_ratio, auto_f0, noise_scale,pad_seconds,cl_num,lg_num,lgr_num,F0_mean_pooling,enhancer_adaptive_key,cr_threshold) File "/root/so-vits-svc4/inference/infer_tool.py", line 285, in slice_inference out_audio, out_sr = self.infer(spk, tran, raw_path, File "/root/so-vits-svc4/inference/infer_tool.py", line 210, in infer audio = self.net_g_ms.infer(c, f0=f0, g=sid, uv=uv, predict_f0=auto_predict_f0, noice_scale=noice_scale)[0,0].data.float() File "/root/so-vits-svc4/models.py", line 409, in infer x = self.pre(c) * x_mask + self.emb_uv(uv.long()).transpose(1,2) File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1190, in _call_impl return forward_call(*input, **kwargs) File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 313, in forward return self._conv_forward(input, self.weight, self.bias) File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 309, in _conv_forward return F.conv1d(input, weight, bias, self.stride, RuntimeError: Given groups=1, weight of size [192, 768, 5], expected input[1, 256, 773] to have 768 channels, but got 256 channels instead
No response
应该是配置文件错误,请使用vev768l12的配置文件
4.1 和 4.0 不兼容,不是开发问题不要用 issues 询问
系统平台版本号
windows 11
GPU 型号
4090
Python版本
3.10
PyTorch版本
n/a
sovits分支
4.0(默认)
数据集来源(用于判断数据集质量)
个人声音处理
出现问题的环节或执行的命令
推理
情况描述
训练2400步后,生成的模型,在预处理的时候报错 异常信息:Given groups=1, weight of size [192, 768, 5], expected input[1, 256, 773] to have 768 channels, but got 256 channels instead 请排障后重试
日志
补充说明
No response