Datasets:
Tasks:
Text2Text Generation
Modalities:
Text
Formats:
parquet
Sub-tasks:
open-domain-abstractive-qa
Languages:
English
Size:
100K - 1M
License:
Commit
•
81631d5
1
Parent(s):
50b3fed
Add QDMR-high-level data files
Browse files
QDMR-high-level/test-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:edf00b230e5f1478bcc55f4e10d751c98d330e732debf5f7885630d3f1ffdbf8
|
3 |
+
size 262748
|
QDMR-high-level/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ea809a3aaeb31982a4a64ef5c07aeab30ef4cd878fcd199a4f10408e429e4f5f
|
3 |
+
size 2412936
|
QDMR-high-level/validation-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4d21ca330865ec917233df3cd6e7f2f3ecf43a7bf8ccfe377635426b27909adb
|
3 |
+
size 437503
|
README.md
CHANGED
@@ -57,17 +57,17 @@ dataset_info:
|
|
57 |
- name: split
|
58 |
dtype: string
|
59 |
splits:
|
60 |
-
- name: test
|
61 |
-
num_bytes: 482339
|
62 |
-
num_examples: 3195
|
63 |
- name: train
|
64 |
-
num_bytes:
|
65 |
num_examples: 17503
|
66 |
- name: validation
|
67 |
-
num_bytes:
|
68 |
num_examples: 3130
|
69 |
-
|
70 |
-
|
|
|
|
|
|
|
71 |
- config_name: QDMR-high-level-lexicon
|
72 |
features:
|
73 |
- name: source
|
@@ -139,6 +139,14 @@ configs:
|
|
139 |
path: QDMR/validation-*
|
140 |
- split: test
|
141 |
path: QDMR/test-*
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
142 |
---
|
143 |
|
144 |
# Dataset Card for "break_data"
|
|
|
57 |
- name: split
|
58 |
dtype: string
|
59 |
splits:
|
|
|
|
|
|
|
60 |
- name: train
|
61 |
+
num_bytes: 5134938
|
62 |
num_examples: 17503
|
63 |
- name: validation
|
64 |
+
num_bytes: 912408
|
65 |
num_examples: 3130
|
66 |
+
- name: test
|
67 |
+
num_bytes: 479919
|
68 |
+
num_examples: 3195
|
69 |
+
download_size: 3113187
|
70 |
+
dataset_size: 6527265
|
71 |
- config_name: QDMR-high-level-lexicon
|
72 |
features:
|
73 |
- name: source
|
|
|
139 |
path: QDMR/validation-*
|
140 |
- split: test
|
141 |
path: QDMR/test-*
|
142 |
+
- config_name: QDMR-high-level
|
143 |
+
data_files:
|
144 |
+
- split: train
|
145 |
+
path: QDMR-high-level/train-*
|
146 |
+
- split: validation
|
147 |
+
path: QDMR-high-level/validation-*
|
148 |
+
- split: test
|
149 |
+
path: QDMR-high-level/test-*
|
150 |
---
|
151 |
|
152 |
# Dataset Card for "break_data"
|
dataset_infos.json
CHANGED
@@ -1,76 +1,63 @@
|
|
1 |
{
|
2 |
"QDMR-high-level": {
|
3 |
-
"description": "Break is a human annotated dataset of natural language questions and their Question Decomposition Meaning Representations\n(QDMRs). Break consists of 83,978 examples sampled from 10 question answering datasets over text, images and databases
|
4 |
"citation": "@article{Wolfson2020Break,\n title={Break It Down: A Question Understanding Benchmark},\n author={Wolfson, Tomer and Geva, Mor and Gupta, Ankit and Gardner, Matt and Goldberg, Yoav and Deutch, Daniel and Berant, Jonathan},\n journal={Transactions of the Association for Computational Linguistics},\n year={2020},\n}\n",
|
5 |
"homepage": "https://github.com/allenai/Break",
|
6 |
"license": "",
|
7 |
"features": {
|
8 |
"question_id": {
|
9 |
"dtype": "string",
|
10 |
-
"id": null,
|
11 |
"_type": "Value"
|
12 |
},
|
13 |
"question_text": {
|
14 |
"dtype": "string",
|
15 |
-
"id": null,
|
16 |
"_type": "Value"
|
17 |
},
|
18 |
"decomposition": {
|
19 |
"dtype": "string",
|
20 |
-
"id": null,
|
21 |
"_type": "Value"
|
22 |
},
|
23 |
"operators": {
|
24 |
"dtype": "string",
|
25 |
-
"id": null,
|
26 |
"_type": "Value"
|
27 |
},
|
28 |
"split": {
|
29 |
"dtype": "string",
|
30 |
-
"id": null,
|
31 |
"_type": "Value"
|
32 |
}
|
33 |
},
|
34 |
-
"
|
35 |
-
"
|
36 |
"config_name": "QDMR-high-level",
|
37 |
"version": {
|
38 |
"version_str": "1.0.0",
|
39 |
-
"description": "",
|
40 |
-
"datasets_version_to_prepare": null,
|
41 |
"major": 1,
|
42 |
"minor": 0,
|
43 |
"patch": 0
|
44 |
},
|
45 |
"splits": {
|
46 |
-
"test": {
|
47 |
-
"name": "test",
|
48 |
-
"num_bytes": 482339,
|
49 |
-
"num_examples": 3195,
|
50 |
-
"dataset_name": "break_data"
|
51 |
-
},
|
52 |
"train": {
|
53 |
"name": "train",
|
54 |
-
"num_bytes":
|
55 |
"num_examples": 17503,
|
56 |
-
"dataset_name":
|
57 |
},
|
58 |
"validation": {
|
59 |
"name": "validation",
|
60 |
-
"num_bytes":
|
61 |
"num_examples": 3130,
|
62 |
-
"dataset_name":
|
63 |
-
}
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
"
|
68 |
-
"
|
69 |
}
|
70 |
},
|
71 |
-
"download_size":
|
72 |
-
"dataset_size":
|
73 |
-
"size_in_bytes":
|
74 |
},
|
75 |
"QDMR-high-level-lexicon": {
|
76 |
"description": "Break is a human annotated dataset of natural language questions and their Question Decomposition Meaning Representations\n(QDMRs). Break consists of 83,978 examples sampled from 10 question answering datasets over text, images and databases. \nThis repository contains the Break dataset along with information on the exact data format.\n",
|
|
|
1 |
{
|
2 |
"QDMR-high-level": {
|
3 |
+
"description": "Break is a human annotated dataset of natural language questions and their Question Decomposition Meaning Representations\n(QDMRs). Break consists of 83,978 examples sampled from 10 question answering datasets over text, images and databases.\nThis repository contains the Break dataset along with information on the exact data format.\n",
|
4 |
"citation": "@article{Wolfson2020Break,\n title={Break It Down: A Question Understanding Benchmark},\n author={Wolfson, Tomer and Geva, Mor and Gupta, Ankit and Gardner, Matt and Goldberg, Yoav and Deutch, Daniel and Berant, Jonathan},\n journal={Transactions of the Association for Computational Linguistics},\n year={2020},\n}\n",
|
5 |
"homepage": "https://github.com/allenai/Break",
|
6 |
"license": "",
|
7 |
"features": {
|
8 |
"question_id": {
|
9 |
"dtype": "string",
|
|
|
10 |
"_type": "Value"
|
11 |
},
|
12 |
"question_text": {
|
13 |
"dtype": "string",
|
|
|
14 |
"_type": "Value"
|
15 |
},
|
16 |
"decomposition": {
|
17 |
"dtype": "string",
|
|
|
18 |
"_type": "Value"
|
19 |
},
|
20 |
"operators": {
|
21 |
"dtype": "string",
|
|
|
22 |
"_type": "Value"
|
23 |
},
|
24 |
"split": {
|
25 |
"dtype": "string",
|
|
|
26 |
"_type": "Value"
|
27 |
}
|
28 |
},
|
29 |
+
"builder_name": "parquet",
|
30 |
+
"dataset_name": "break_data",
|
31 |
"config_name": "QDMR-high-level",
|
32 |
"version": {
|
33 |
"version_str": "1.0.0",
|
|
|
|
|
34 |
"major": 1,
|
35 |
"minor": 0,
|
36 |
"patch": 0
|
37 |
},
|
38 |
"splits": {
|
|
|
|
|
|
|
|
|
|
|
|
|
39 |
"train": {
|
40 |
"name": "train",
|
41 |
+
"num_bytes": 5134938,
|
42 |
"num_examples": 17503,
|
43 |
+
"dataset_name": null
|
44 |
},
|
45 |
"validation": {
|
46 |
"name": "validation",
|
47 |
+
"num_bytes": 912408,
|
48 |
"num_examples": 3130,
|
49 |
+
"dataset_name": null
|
50 |
+
},
|
51 |
+
"test": {
|
52 |
+
"name": "test",
|
53 |
+
"num_bytes": 479919,
|
54 |
+
"num_examples": 3195,
|
55 |
+
"dataset_name": null
|
56 |
}
|
57 |
},
|
58 |
+
"download_size": 3113187,
|
59 |
+
"dataset_size": 6527265,
|
60 |
+
"size_in_bytes": 9640452
|
61 |
},
|
62 |
"QDMR-high-level-lexicon": {
|
63 |
"description": "Break is a human annotated dataset of natural language questions and their Question Decomposition Meaning Representations\n(QDMRs). Break consists of 83,978 examples sampled from 10 question answering datasets over text, images and databases. \nThis repository contains the Break dataset along with information on the exact data format.\n",
|