ignorejjj / MetaRAG

Code for the paper: Metacognitive Retrieval-Augmented Large Language Models
MIT License
13 stars 3 forks source link

Unrecognized index name wikipedia-dpr-100w #1

Closed 2282588541a closed 1 month ago

2282588541a commented 1 month ago

I try run this code in my local environment, but when run the code likefrom pyserini.search.lucene import LuceneSearcher searcher = LuceneSearcher.from_prebuilt_index('wikipedia-dpr-100w') The result show Unrecognized index name wikipedia-dpr-100w. I want to know how I can slove this

ignorejjj commented 1 month ago

See this: https://github.com/castorini/pyserini/blob/master/pyserini/resources/index-metadata/index-wikipedia-dpr-20210120-d1b9e6-readme.txt

2282588541a commented 1 month ago

emm,I have try this ,It looks like I have to use anserinito generate index. When I apply anserini ,there are too many problems(the meaven show many errrors and I can't solve them), If conditions permit, can you upload the index you built? Aliyun, Google Drive. Thank you very much. Or will it have a big impact on the results if I replace the search part with other BM25-based search engines (similar to elastic)?

ignorejjj commented 1 month ago

You can download it from pyserini: https://github.com/castorini/pyserini/blob/master/docs/usage-interactive-search.md#how-do-i-manually-download-indexes

2282588541a commented 1 month ago

thank you very much!!!Anyone can download this corpus in this link:https://git.uwaterloo.ca/jimmylin/anserini-indexes/raw/master/index-wikipedia-dpr-20210120-d1b9e6.tar.gz