forked from castorini/anserini
-
Notifications
You must be signed in to change notification settings - Fork 0
/
msmarco-v1-passage.splade-pp-sd.yaml
94 lines (89 loc) · 2.35 KB
/
msmarco-v1-passage.splade-pp-sd.yaml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
---
corpus: msmarco-passage-splade-pp-sd
corpus_path: collections/msmarco/msmarco-passage-splade-pp-sd
download_url: https://rgw.cs.uwaterloo.ca/pyserini/data/msmarco-passage-splade-pp-sd.tar
download_checksum: cb7e264222f2bf2221dd2c9d28190be1
index_path: indexes/lucene-inverted.msmarco-v1-passage.splade-pp-sd/
collection_class: JsonVectorCollection
generator_class: DefaultLuceneDocumentGenerator
index_threads: 16
index_options: -impact -pretokenized -storeDocvectors
index_stats:
documents: 8841823
documents (non-empty): 8841823
total terms: 55456660129
metrics:
- metric: AP@1000
command: bin/trec_eval
params: -c -m map
separator: "\t"
parse_index: 2
metric_precision: 4
can_combine: false
- metric: RR@10
command: bin/trec_eval
params: -c -M 10 -m recip_rank
separator: "\t"
parse_index: 2
metric_precision: 4
can_combine: false
- metric: R@100
command: bin/trec_eval
params: -c -m recall.100
separator: "\t"
parse_index: 2
metric_precision: 4
can_combine: false
- metric: R@1000
command: bin/trec_eval
params: -c -m recall.1000
separator: "\t"
parse_index: 2
metric_precision: 4
can_combine: false
topic_reader: TsvInt
topics:
- name: "[MS MARCO Passage: Dev](https://github.com/microsoft/MSMARCO-Passage-Ranking)"
id: dev
path: topics.msmarco-passage.dev-subset.splade-pp-sd.tsv.gz
qrel: qrels.msmarco-passage.dev-subset.txt
models:
- name: splade-pp-sd-cached_q
display: SPLADE++ CoCondenser-SelfDistil
params: -impact -pretokenized
results:
AP@1000:
- 0.3837
RR@10:
- 0.3778
R@100:
- 0.9118
R@1000:
- 0.9846
# PRF regressions are no longer maintained for sparse judgments to reduce running times.
# (commenting out instead of removing; in case these numbers are needed, just uncomment and rerun.)
#
# - name: splade-pp-sd-cached_q+rm3
# display: +RM3
# params: -impact -pretokenized -rm3
# results:
# AP@1000:
# - 0.2879
# RR@10:
# - 0.2790
# R@100:
# - 0.8681
# R@1000:
# - 0.9739
# - name: splade-pp-sd-cached_q+rocchio
# display: +Rocchio
# params: -impact -pretokenized -rocchio
# results:
# AP@1000:
# - 0.3350
# RR@10:
# - 0.3279
# R@100:
# - 0.8943
# R@1000:
# - 0.9826