seungwonpark / melgan

MelGAN vocoder (compatible with NVIDIA/tacotron2)
http://swpark.me/melgan/
BSD 3-Clause "New" or "Revised" License
636 stars 116 forks source link

RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same #70

Open xjsdn opened 1 year ago

xjsdn commented 1 year ago

2022-11-05 01:29:45,173 - INFO - Starting new training run. Validation loop: 0%| | 0/710 [00:00<?, ?it/s]<datasets.dataloader.MelFromDisk object at 0x000002EAEF66B710> tensor([[ -6.0881, -6.2532, -6.2991, ..., -11.5129, -11.5129, -11.5129], [ -6.9093, -5.9843, -5.8405, ..., -11.5129, -11.5129, -11.5129], [ -6.8329, -6.0982, -6.6101, ..., -11.5129, -11.5129, -11.5129], ..., [ -8.3635, -8.1427, -8.2876, ..., -11.5129, -11.5129, -11.5129], [ -8.0431, -7.8149, -8.1762, ..., -11.5129, -11.5129, -11.5129], [ -7.9314, -8.1326, -8.2959, ..., -11.5129, -11.5129, -11.5129]]) Validation loop: 0%| | 0/710 [00:00<?, ?it/s] 2022-11-05 01:29:45,357 - INFO - Exiting due to exception: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same Traceback (most recent call last): File "E:\ai_ai\melgan-master\utils\train.py", line 60, in train validate(hp, args, model_g, model_d, valloader, writer, step) File "E:\ai_ai\melgan-master\utils\validation.py", line 23, in validate fake_audio = generator(mel) File "E:\ai_ai\melgan-master\venv\lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl result = self.forward(*input, kwargs) File "E:\ai_ai\melgan-master\model\generator.py", line 48, in forward return self.generator(mel) File "E:\ai_ai\melgan-master\venv\lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl result = self.forward(*input, *kwargs) File "E:\ai_ai\melgan-master\venv\lib\site-packages\torch\nn\modules\container.py", line 119, in forward input = module(input) File "E:\ai_ai\melgan-master\venv\lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl result = self.forward(input, kwargs) File "E:\ai_ai\melgan-master\venv\lib\site-packages\torch\nn\modules\conv.py", line 263, in forward return self._conv_forward(input, self.weight, self.bias) File "E:\ai_ai\melgan-master\venv\lib\site-packages\torch\nn\modules\conv.py", line 260, in _conv_forward self.padding, self.dilation, self.groups) RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same