Training and prediction on new dataset

[akshayayadav]

Hey plkmo,

I was wondering if the main_task.py could be used to train on a completely new set of relationships different from the semeval dataset. Could you please let me know the changes that I need to make in the code to train and predict on new data?

Thanks

[demongolem-biz]

@plkmo I am in the same situation. I have my own ontology will spells out a certain set of relation types. Will I have to go back to the main_pretraining.py file and start over again?

[TahaAslani]

I am trying to do the same thing, but I cannot tailor the code for fine-tuning on a new set.

[plkmo]

Hi all, if you need to fine-tune on your own dataset other than Semeval, you will have to fork this and modify src/tasks/preprocessing_funcs.py to ingest your own data and reformat it similar to semeval dataset format.

When I have time, I will implement this feature