import pyterrier as pt
index = pt.Artifact.from_hf('pyterrier/nq.terrier')
index.bm25()
Benchmarks
name
nDCG@10
R@1000
bm25
0.2814
0.8906
dph
0.2846
0.8926
Reproduction
import pyterrier as pt
from tqdm import tqdm
import pandas as pd
import ir_datasets
from pyterrier_pisa import PisaIndex
dataset = ir_datasets.load('beir/nq')… See the full description on the dataset page: https://huggingface.co/datasets/pyterrier/nq.terrier.
