Sugeeth

Results 6 comments of Sugeeth

This is the best I could think of 1. In [trainer.py](https://github.com/Raghava14/OpenNMT-py/blob/best_checkpoint/onmt/trainer.py#L276) added force validation at every save_checkpoint_steps. 2. While saving the model if it is triggered by early_stopping best step...

@francoishernandez any update on this ?

Me too facing the same issue using T2T :1.6.3. Getting same as mentioned in log by @Qiaoxl . Any help on this @rsepassi ??

@rsepassi can I use sentencepiece to generate vocabulary and keep it in t2t_data folder ?? in that case can I simply run the t2t_trainer or any other changes are needed?...

Thanks for the reply. Please update when it is available. I am trying to export` transformer_base` model currently to Caffe2.

Hi @aastha19 no I didn't make any progress as of now I will share if I do. Also please let me know if you make any. Thanks.