castorini / pyserini

Pyserini is a Python toolkit for reproducible information retrieval research with sparse and dense representations.

Home Page:http://pyserini.io/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

`castorini/unicoil-d2q-msmarco-passage` referenced in documentation is missing

zeuner opened this issue · comments

encoder --encoder castorini/unicoil-d2q-msmarco-passage \

Is there a way to recreate the model, or a different model to be used instead in the example for encoding documents with the sparse encoder?

Hi @zeuner, the model is this one, https://huggingface.co/castorini/unicoil-msmarco-passage. This model is trained with d2q augmentation.

There is also a model without the doc2query augumentation for the corpus: https://huggingface.co/castorini/unicoil-noexp-msmarco-passage.