Jimmy Lin

> Fixed the typo that caused this evaluation command to fail for bge-base-en-v1.5 cached queries: > > ``` > % java -cp /Users/jimmylin/workspace/anserini/target/anserini-0.36.1-SNAPSHOT-fatjar.jar io.anserini.search.SearchCollection -threads 16 -index beir-v1.0.0-trec-covid.bge-base-en-v1.5 -topics beir-trec-covid.bge-base-en-v1.5...

Another thing to consider: what does Elasticsearch and OpenSearch expose? One option is just to mirror that?

First impl here #2486 - @16BitNarwhal do you want to leave this issue open or close it?

> I think the former is more efficient in terms of memory (and a more natural view), but it does require more thought and possibly more work in terms of...

Okay, I'll send me GitLab, etc.

The duobert data takes up 69G, which is a non-trivial amount of space to be paying for and holding on GCP forever, so this is good.

Archived data from GCP to Dropbox/Gitlab in #9 - Let's _not_ change into read-only for a bit - wait until things settle down a bit just in case we notice...

Can you provide more details please?

Please consult our paper here: https://arxiv.org/abs/2102.13019

hey @joelrorseth thanks for noting these issues. I just pinged you on Slack as follow-up. I don't think anyone has tried to replicate this for a while, so it's great...