You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Mar 1, 2024. It is now read-only.
Hi BLINK's authors,
Thanks for releasing the pretrained model and the code! They are very useful.
I just wonder where I can find the training script to train the bi-encoder, cross-encoder models on my own data.
@vinkhuc: It is indeed possible to train a model on your own custom dataset. I would recommend first training a zeshel (Zero-shot learning) model, then bring your data in the same format as zeshel and train your model.
Scripts to download and pre-process zeshel data set are in the repository.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Hi BLINK's authors,
Thanks for releasing the pretrained model and the code! They are very useful.
I just wonder where I can find the training script to train the bi-encoder, cross-encoder models on my own data.
ELQ has the training script available at: https://github.com/facebookresearch/BLINK/tree/master/elq_slurm_scripts
Could anyone point me to the corresponding training script for BLINK? I'm looking at the entity linking task for long texts.
Thanks in advance!
The text was updated successfully, but these errors were encountered: