Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

2CR for dense AToMiC retrieval #1674

Merged
merged 9 commits into from
Oct 12, 2023
Merged

Conversation

jasper-xian
Copy link
Member

@jasper-xian jasper-xian commented Oct 8, 2023

AToMiC dense 2CR for model ViT-L-14.laion2b_s32b_b82k

Related PRs: castorini/anserini#2219 castorini/anserini-tools#57 castorini/pyserini-data#6

@jasper-xian jasper-xian marked this pull request as draft October 8, 2023 01:47
@jasper-xian
Copy link
Member Author

jasper-xian commented Oct 8, 2023

Hey @lintool, am looking to add some topic files to anserini so we can do a 2CR. These are json files that follow the DefaultQueryIterator format, so basically:

{"<id>": {"title": "<some text>"}, ....}

Which anserini TopicReader would work for this sort of topic file? Let me know and I'll add a corresponding PR in anserini as well.

For reference, I'm following this PR as a guide on how to add topics to anserini.

@lintool
Copy link
Member

lintool commented Oct 8, 2023

@jasper-xian jasper-xian marked this pull request as ready for review October 11, 2023 16:13
Copy link
Member

@justram justram left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good. Just need to change a few names to align the paper:

  1. Eval settings: validation -> small
  2. A column to show the query set we used in the table: val queries

pyserini/2cr/atomic.py Outdated Show resolved Hide resolved
pyserini/2cr/atomic.py Outdated Show resolved Hide resolved
pyserini/2cr/atomic_html.template Show resolved Hide resolved
@jasper-xian
Copy link
Member Author

changes made as requested!

Copy link
Member

@justram justram left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@jasper-xian jasper-xian merged commit f889bc4 into castorini:master Oct 12, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants