Fine Tuning methodology.

#1
by reza-takhshid - opened

Hey @Crystalcareai
I've been trying to fine-tune the LargeV3, but I see weird behaviours like "forgetting" or inconsistent changes in loss/WER during training.

Can you share your training script or hyperparameters. Or if you have any hints about this.
Thanks

Sign up or log in to comment