Training the a new model based on previous trained model

[siddhant sharma]

Hi All,

Currently I am training a text-text translation model, which is generating the checkpoint files which are in turn the models in the the output_dir location.

Now if I want to start training some other data-set again using the transformer model but want the training to start over the previously trained model, then what should I do?

Reverences 
Siddhant Sharma

[Martin Popel]

Just use the same output_dir and continue training.
Make sure the new dataset uses the same subword vocabulary as the original training data.

Martin

> *Reverences*
> *Siddhant Sharma*
>
> --
> You received this message because you are subscribed to the Google Groups
> "tensor2tensor" group.
> To unsubscribe from this group and stop receiving emails from it, send an email
> to tensor2tenso...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/tensor2tensor/CAMQW_1ASNzcPLbM%3DrBcCn0AsfkHhMgmOTZQry2v00LHuvn8ksA%40mail.gmail.com.