Training on own model

[ciuhc]

Hi, amazing work all!
From what I understand it is possible to train the model on my own data. I've got a sound library, which I want to use to train the model. Do I just replace the .json elements in the data folder?

/Update:
I replaced the data folder json's with mine and after some back and forth I'm stick at the following error. Of course the columns match:

ValueError: Couldn't cast
dataset: string
location: string
to
{'dataset': Value(dtype='string', id=None), 'location': Value(dtype='string', id=None), 'captions': Value(dtype='string', id=None)}
because column names don't match

Perhapse the issue is how the file is made? I pulled he metadata into xlsx -> made python script to make json and did some formatting fixes.

thx!

Ps. how big is your training setup in terms of GPU's?

[deepanwayx]

ValueError: Couldn't cast
dataset: string
location: string

This doesn't have the captions field. That's why the column names do not match. Perhaps you could use some metadata about your audio clips as captions? TANGO needs some text captions as input on which it conditions the generated outputs.

We used a 4* RTX A 6000 for training.

[ciuhc]

Seems like it was having issues with the json file - when I made a csv it worked fine. Probably something with the json creation script.
Thanks!