The cause of mismatch for model params

[azraelkuan]

for your note:

Our UnivNet generator has smaller number of parameters (c32: 5.11M, c16: 1.42M) than the paper (c32: 14.89M, c16: 4.00M). So far, we have not encountered any issues from using a smaller model size. If run into any problem, please report it as an issue.

it should be a mistake in your code:
https://github.com/mindslab-ai/univnet/blob/df77c9a37f71e3d6be1b504e16abaf99ce131de3/model/lvcnet.py#L110

the default value should be 3, but u set to 1 in the code.

[wookladin]

Thanks for reporting the problem!
We really appreciate it.
As many open sources do, we want to invite you as a contributor to this repository.
To become a contributor, you should fork this repository, edit it, and send us a pull request.
We will accept your pull request when training of our modified model is done.
If you don't want to be a contributor, you don't have to do this.

Thanks again.

[azraelkuan]

@wookladin i have created a pr, so close this.