ranx icon indicating copy to clipboard operation
ranx copied to clipboard

⚡️A Blazing-Fast Python Library for Ranking Evaluation, Comparison, and Fusion 🐍

Results 10 ranx issues
Sort by recently updated
recently updated
newest added

**Is your feature request related to a problem? Please describe.** I think it’s quite frustrating to have to specify the format of qrels/run (TREC or JSON). I often get exceptions...

enhancement

I understood that, when evaluating MAP@k, relevance judgment scores equal to 0 are ignored. In my case, I get a bit of a weird behaviour. I'm working on a balanced...

Hi! This lib is extremely good tool to have in arsenal, but I think it could be nice to have Spearman and Kendall correlation functions included to this package to...

enhancement

I'm noticing that it takes a really long time to start up anything with ranx in a serverless setting, e.g. Modal Labs. I tried adding a step during the build...

help wanted

If the retriever is operating at passage / chunk-level and hence the retrieved results can have duplicate IDs as shown below: Top-10 retrieval list for each of the queries: ```...

Hi ! I've created a conversion to DataFrame with to_dataframe() for the report outputs. I believe this will be useful for those who want the report results in numeric form...

I came across a point in the [Fuhr](https://www.sigir.org/wp-content/uploads/2018/01/p032.pdf) article where it talks about the need for additional correction for multiple hypotheses test (e.g., Bonferroni correction). Looking at the `paired_student_t_test` function...

How do we compare different runs with multiple folds per run? For instance, assume we have 10-folds for `run_1`, ... `run_5`? ```python from ranx import compare # Compare different runs...

**Is your feature request related to a problem? Please describe.** In some extremely challenging scenarios for information retrieval, items relevant to a query follow a long tail distribution. Thus, there...

enhancement

## Background Introduction: I need to perform a hybrid search using **ranx** and [**MyScaleDB**](https://github.com/myscale/myscaledb). The principle is as follows: - Use MyScale's `TextSearch` to obtain the `BM25` document scores (higher...