macavaney/msmarco-passage.anserini

MS MARCO Anserini Index

Description

This is an index of the MS MARCO passage (v1) dataset with Anserini. It can be used for passage retrieval using lexical methods.

Usage

>>> from pyterrier_anserini import AnseriniIndex
>>> index = AnseriniIndex.from_hf('macavaney/msmarco-passage.anserini')
>>> bm25 = index.bm25(include_fields=['contents'])
>>> bm25.search('terrier breeds')
  qid           query    docno    score  rank                                      contents
0   1  terrier breeds  5785957  11.9588     0  The Jack Russell Terrier and the Russell ...
1   1  terrier breeds  7455374  11.9343     1  FCI, ANKC, and IKC recognize the shorts a...
2   1  terrier breeds  1406578  11.8640     2  Norfolk terrier (English breed of small t...
3   1  terrier breeds  3984886  11.7518     3  Terrier Group is the name of a breed Grou...
4   1  terrier breeds  7728131  11.5660     4  The Yorkshire Terrier didn't begin as the...
...

Benchmarks

TREC DL 2019

Code

from ir_measures import nDCG, RR, MAP, R
import pyterrier as pt
from pyterrier_anserini import AnseriniIndex
index = AnseriniIndex.from_hf('macavaney/msmarco-passage.anserini')
dataset = pt.get_dataset('irds:msmarco-passage/trec-dl-2019/judged')
pt.Experiment(
  [index.bm25(), index.qld(), index.tfidf()],
  dataset.get_topics(),
  dataset.get_qrels(),
  [nDCG@10, nDCG, RR(rel=2), MAP(rel=2), R(rel=2)@1000],
  ['BM25', 'QLD', 'TF-IDF'],
  round=4,
)

	name	nDCG@10	nDCG	RR(rel=2)	AP(rel=2)	R(rel=2)@1000
0	BM25	0.5121	0.61	0.715	0.3069	0.7529
1	QLD	0.4689	0.5995	0.606	0.3014	0.7662
2	TF-IDF	0.3742	0.5083	0.5203	0.2012	0.7016

TREC DL 2020

Code

from ir_measures import nDCG, RR, MAP, R
import pyterrier as pt
from pyterrier_anserini import AnseriniIndex
index = AnseriniIndex.from_hf('macavaney/msmarco-passage.anserini')
dataset = pt.get_dataset('irds:msmarco-passage/trec-dl-2020/judged')
pt.Experiment(
  [index.bm25(), index.qld(), index.tfidf()],
  dataset.get_topics(),
  dataset.get_qrels(),
  [nDCG@10, nDCG, RR(rel=2), MAP(rel=2), R(rel=2)@1000],
  ['BM25', 'QLD', 'TF-IDF'],
  round=4,
)

	name	nDCG@10	nDCG	RR(rel=2)	AP(rel=2)	R(rel=2)@1000
0	BM25	0.4769	0.5832	0.672	0.2827	0.7865
1	QLD	0.4584	0.5872	0.6238	0.2811	0.8179
2	TF-IDF	0.4029	0.5039	0.5526	0.2107	0.7323

MS MARCO Dev (small)

Code

from ir_measures import RR, R
import pyterrier as pt
from pyterrier_anserini import AnseriniIndex
index = AnseriniIndex.from_hf('macavaney/msmarco-passage.anserini')
dataset = pt.get_dataset('irds:msmarco-passage/dev/small')
pt.Experiment(
  [index.bm25(), index.qld(), index.tfidf()],
  dataset.get_topics(),
  dataset.get_qrels(),
  [RR@10, R@1000],
  ['BM25', 'QLD', 'TF-IDF'],
  round=4,
)

	name	RR@10	R@1000
0	BM25	0.1844	0.8567
1	QLD	0.1664	0.8508
2	TF-IDF	0.1368	0.8288

Reproduction

>>> import pyterrier as pt
>>> import pyterrier_anserini
>>> idx = pyterrier_anserini.AnseriniIndex('msmarco-passage.anserini')
>>> idx.indexer().index(pt.get_dataset('irds:msmarco-passage').get_corpus_iter())

Metadata

{
  "type": "sparse_index",
  "format": "anserini",
  "package_hint": "pyterrier-anserini"
}