search

NVIDIA / mellotron

Mellotron: a multispeaker voice synthesis model based on Tacotron 2 GST that can make a voice emote and sing without emotive or singing training data
BSD 3-Clause "New" or "Revised" License
853 stars 187 forks source link

colab error #98

Open mochacat2001 opened 3 years ago

mochacat2001 commented 3 years ago

Train loss 55 0.477043 Grad Norm 0.824708 4.30s/it here Traceback (most recent call last): File "train.py", line 286, in args.warm_start, args.n_gpus, args.rank, args.group_name, hparams) File "train.py", line 210, in train y_pred = model(x) File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl result = self.forward(*input, kwargs) File "/content/mellotron/model.py", line 611, in forward embedded_text = self.encoder(embedded_inputs, input_lengths) File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl result = self.forward(*input, *kwargs) File "/content/mellotron/model.py", line 193, in forward curr_x = F.dropout(F.relu(conv(curr_x)), drop_rate, self.training) File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl result = self.forward(input, kwargs) File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/container.py", line 119, in forward input = module(input) File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl result = self.forward(*input, **kwargs) File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/batchnorm.py", line 140, in forward self.weight, self.bias, bn_training, exponential_average_factor, self.eps) File "/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py", line 2147, in batch_norm _verify_batch_size(input.size()) File "/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py", line 2114, in _verify_batch_size raise ValueError("Expected more than 1 value per channel when training, got input size {}".format(size)) ValueError: Expected more than 1 value per channel when training, got input size torch.Size([1, 512, 1])

josebeo2016 commented 2 years ago

I got the same issue in the same train loss 55

BlazJurisic commented 1 year ago

anyone resolved this issue?