We evaluate our models using pytrec_eval and in future we can extend to include more retrieval-based metrics:
NDCG@k
MAP@k
Recall@k
P@k
We also include custom-metrics now which can be used for evaluation, please refer here - evaluate_custom_metrics.py
MRR@k
R_cap@k
Hole@k
Accuracy@k
There was an error while loading. Please reload this page.
Home
Installing BEIR
Examples and Tutorials
Datasets Available
Multilingual Datasets
Load your Custom Dataset
Models Available
Evaluate your Custom Model
Metrics Available
Leaderboard
Couse Material on IR