⚡️ Introduction

ranx ([raŋks]) is a library of fast ranking evaluation metrics implemented in Python, leveraging Numba for high-speed vector operations and automatic parallelization. It offers a user-friendly interface to evaluate and compare Information Retrieval and Recommender Systems. ranx allows you to perform statistical tests and export LaTeX tables for your scientific publications. Moreover, ranx provides several fusion algorithms and normalization strategies, and an automatic fusion optimization functionality. ranx also have a companion repository of pre-computed runs to facilitated model comparisons called ranxhub. On ranxhub, you can download and share pre-computed runs for Information Retrieval datasets, such as MSMARCO Passage Ranking. ranx was featured in ECIR 2022, CIKM 2022, and SIGIR 2023.

If you use ranx to evaluate results or conducting experiments involving fusion for your scientific publication, please consider citing it: evaluation bibtex, fusion bibtex, ranxhub bibtex.

NB: ranx is not suited for evaluating classifiers. Please, refer to the FAQ for further details.

For a quick overview, follow the Usage section.

For a in-depth overview, follow the Examples section.

✨ Features

Metrics

• Hits
• Hit Rate
• Precision
• Recall
• F1
• r-Precision
• Bpref
• Rank-biased Precision (RBP)
• Mean Reciprocal Rank (MRR)
• Mean Average Precision (MAP)
• Discounted Cumulative Gain (DCG)
• Normalized Discounted Cumulative Gain (NDCG)

The metrics have been tested against TREC Eval for correctness.

Statistical Tests

• Paired Student's t-Test (default)
• Fisher's Randomization Test
• Tukey's HSD Test

Please, refer to Smucker et al., Carterette, and Fuhr for additional information on statistical tests for Information Retrieval.

Off-the-shelf Qrels

You can load qrels from ir-datasets as simply as: python qrels = Qrels.from_ir_datasets("msmarco-document/dev") A full list of the available qrels is provided here.

Off-the-shelf Runs

You can load runs from ranxhub as simply as: python run = Run.from_ranxhub("run-id") A full list of the available runs is provided here.

Fusion Algorithms

| Name | Name | Name | Name | Name | | -------------------------------------------------------- | ---------------------------------------------------------- | ----------------------------------------------------------------------- | ------------------------------------------------------------ | ------------------------------------------------------------------------------ | | CombMIN | CombMNZ | RRF | MAPFuse | BordaFuse | | CombMED | CombGMNZ | RBC | PosFuse | Weighted BordaFuse | | CombANZ | ISR | WMNZ | ProbFuse | Condorcet | | CombMAX | Log_ISR | Mixed | SegFuse | Weighted Condorcet | | CombSUM | LogN_ISR | BayesFuse | SlideFuse | Weighted Sum |

Please, refer to the documentation for further details.

Normalization Strategies

• Min-Max Norm
• Min-Max Inverted Norm
• Max Norm
  
• Sum Norm
  
• ZMUV Norm
  
• Rank Norm
  
• Borda Norm

Please, refer to the documentation for further details.

🔌 Requirements

bash python>=3.8 As of v.0.3.5, ranx requires python>=3.8.

💾 Installation

bash pip install ranx

💡 Usage

Create Qrels and Run

```python from ranx import Qrels, Run

qrelsdict = { "q1": { "d12": 5, "d25": 3 }, "q2": { "d11": 6, "d_22": 1 } }

rundict = { "q1": { "d12": 0.9, "d23": 0.8, "d25": 0.7, "d36": 0.6, "d32": 0.5, "d35": 0.4 }, "q2": { "d12": 0.9, "d11": 0.8, "d25": 0.7, "d36": 0.6, "d22": 0.5, "d_35": 0.4 } }

qrels = Qrels(qrelsdict) run = Run(rundict) ```

Evaluate

```python from ranx import evaluate

Compute score for a single metric

evaluate(qrels, run, "ndcg@5")

0.7861

Compute scores for multiple metrics at once

evaluate(qrels, run, ["map@5", "mrr"])

{"map@5": 0.6416, "mrr": 0.75} ```

Compare

```python from ranx import compare

Compare different runs and perform Two-sided Paired Student's t-Test

report = compare( qrels=qrels, runs=[run1, run2, run3, run4, run5], metrics=["map@100", "mrr@100", "ndcg@10"], maxp=0.01 # P-value threshold ) Output: python print(report)

Model MAP@100 MRR@100 NDCG@10

a model1 0.320ᵇ 0.320ᵇ 0.368ᵇᶜ b model2 0.233 0.234 0.239 c model3 0.308ᵇ 0.309ᵇ 0.330ᵇ d model4 0.366ᵃᵇᶜ 0.367ᵃᵇᶜ 0.408ᵃᵇᶜ e model_5 0.405ᵃᵇᶜᵈ 0.406ᵃᵇᶜᵈ 0.451ᵃᵇᶜᵈ ```

Fusion

```python from ranx import fuse, optimize_fusion

bestparams = optimizefusion( qrels=trainqrels, runs=[trainrun1, trainrun2, trainrun_3], norm="min-max", # The norm. to apply before fusion method="wsum", # The fusion algorithm to use (Weighted Sum) metric="ndcg@100", # The metric to maximize )

combinedtestrun = fuse( runs=[testrun1, testrun2, testrun3],
norm="min-max",
method="wsum",
params=best_params, ) ```

📖 Examples

| Name | Link | | ---------------------------------------------------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Overview | | | Qrels and Run | | | Evaluation | | | Comparison and Report | | | Fusion | | | Plot | | | Share your runs with ranxhub | |

📚 Documentation

Browse the documentation for more details and examples.

🎓 Citation

If you use ranx to evaluate results for your scientific publication, please consider citing our ECIR 2022 paper:

If you use the fusion functionalities provided by ranx for conducting the experiments of your scientific publication, please consider citing our CIKM 2022 paper:

If you use pre-computed runs from ranxhub to make comparison for your scientific publication, please consider citing our SIGIR 2023 paper:

🎁 Feature Requests

Would you like to see other features implemented? Please, open a feature request.

🤘 Want to contribute?

Would you like to contribute? Please, drop me an e-mail.

📄 License

ranx is an open-sourced software licensed under the MIT license.