I am looking to fine-tune the AnimateDiff model with the dataset I've collected.
The "name" variable for my video data is composed of multiple words or phrases, rather than a single clear sentence.
Is it okay to train the model without preprocessing this data?
Hello!
I am looking to fine-tune the AnimateDiff model with the dataset I've collected. The "name" variable for my video data is composed of multiple words or phrases, rather than a single clear sentence. Is it okay to train the model without preprocessing this data?
Thank you!