Error on tensor dimension
Hello!
I was trying to fine-tune the model for italian language considering the reference on the official page, however when training I get this error:
"RuntimeError: Sizes of tensors must match except in dimension 2. Expected size 16 but got size 256 for tensor number 1 in the list."
Have you encountered the same problem? If so, how did you solve it?
Thank you!
Hi @SaladSlayer00 !
I think I have also gotten such an error at some point and usually it had to do with the way you construct the data (input and output) that you pass to fit the model. Obviously there is some dimension mismatch during model fitting, but you will have to check in your code to understand where exactly it originates from. Personally, I started my journey based on the HuggingFace Audio course (https://huggingface.co/learn/audio-course/chapter0/introduction). You can have a look in the respective section of that course and check also their code template for further training text-to-speech models. It could be of help.