Hello!
First of all, thank you for releasing the source code for TOTO. It is a very impressive model!
Is it possible to share a small example training the backbone model?
I have been trying to reverse engineer an example (using the hyperparameters from your well-written paper) but unfortunately I think I am doing something wrong with the composite loss function.
Thank you in advance!
With kind regards,
Kostas
Hello!
First of all, thank you for releasing the source code for TOTO. It is a very impressive model!
Is it possible to share a small example training the backbone model?
I have been trying to reverse engineer an example (using the hyperparameters from your well-written paper) but unfortunately I think I am doing something wrong with the composite loss function.
Thank you in advance!
With kind regards,
Kostas