-
When attempting to fetch a raw document using the |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 2 replies
-
|
Hi @waylight3 this is the expected behavior. This is because once a document is encoded by SPLADE, it is no longer human-readable, so we've decided not to store them in the prebuilt indexes. Getting the actual human-readable text from the flat index is indeed the preferred approach. |
Beta Was this translation helpful? Give feedback.
-
|
Thank you for the kind answer! Then, can I fecth SPLADE encodings from this index? |
Beta Was this translation helpful? Give feedback.
-
|
You can download the SPLADE encodings from here: https://github.com/castorini/anserini?tab=readme-ov-file#%EF%B8%8F-end-to-end-regression-experiments under "BEIR (v1.0.0) Regressions" tab. |
Beta Was this translation helpful? Give feedback.
You can download the SPLADE encodings from here: https://github.com/castorini/anserini?tab=readme-ov-file#%EF%B8%8F-end-to-end-regression-experiments under "BEIR (v1.0.0) Regressions" tab.