-
Notifications
You must be signed in to change notification settings - Fork 24
vector indexing gt modified #45
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
base: main
Are you sure you want to change the base?
Conversation
@AmberLJC could you double check that the ground truth makes sense? Thanks! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Left some comments. Let's focus on LLM reasoning 1, which reviewer cares about most!
benchmark/experimentation_bench/vector_index/ground_truth/q11.txt
Outdated
Show resolved
Hide resolved
1. **Effect on Increasing Data Size**: ? | ||
2. **Effect on Increasing Mean**: The recall is not significantly affected by changes in the mean, showing robustness. | ||
3. **Effect on Increasing Variance**: The recall remains mostly stable, with some rare instances of decreased performance. | ||
3. **Effect on Increasing Variance**: Higher variance leads to a decline in recall. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think the previous ground truth is more correct
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
When the variance increases, the vectors become more dispersed, making it harder to capture all neighboring points, and the recall rate decreases. Does this reasoning make sense?
Hey @yiliazhuxyxy could you please review the comments left by @AmberLJC ? Thank you! |
This PR is duplicated with #46 |
name: Pull Request
Description
Expected Behavior
Performance Evaluation
This PR has been tested using the first quick start example, with the following performance metrics: