Commit
•
45d4cbe
1
Parent(s):
320355e
Convert dataset to Parquet
Browse filesConvert dataset to Parquet.
- README.md +14 -5
- dataset_infos.json +199 -1
- v1.1/test-00000-of-00001.parquet +3 -0
- v1.1/train-00000-of-00001.parquet +3 -0
- v1.1/validation-00000-of-00001.parquet +3 -0
README.md
CHANGED
|
@@ -26,16 +26,16 @@ dataset_info:
|
|
| 26 |
sequence: string
|
| 27 |
splits:
|
| 28 |
- name: validation
|
| 29 |
-
num_bytes:
|
| 30 |
num_examples: 10047
|
| 31 |
- name: train
|
| 32 |
-
num_bytes:
|
| 33 |
num_examples: 82326
|
| 34 |
- name: test
|
| 35 |
-
num_bytes:
|
| 36 |
num_examples: 9650
|
| 37 |
-
download_size:
|
| 38 |
-
dataset_size:
|
| 39 |
- config_name: v2.1
|
| 40 |
features:
|
| 41 |
- name: answers
|
|
@@ -68,6 +68,15 @@ dataset_info:
|
|
| 68 |
num_examples: 101092
|
| 69 |
download_size: 1384271865
|
| 70 |
dataset_size: 4287455242
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 71 |
---
|
| 72 |
|
| 73 |
# Dataset Card for "ms_marco"
|
|
|
|
| 26 |
sequence: string
|
| 27 |
splits:
|
| 28 |
- name: validation
|
| 29 |
+
num_bytes: 42665198
|
| 30 |
num_examples: 10047
|
| 31 |
- name: train
|
| 32 |
+
num_bytes: 350516260
|
| 33 |
num_examples: 82326
|
| 34 |
- name: test
|
| 35 |
+
num_bytes: 40977580
|
| 36 |
num_examples: 9650
|
| 37 |
+
download_size: 217328153
|
| 38 |
+
dataset_size: 434159038
|
| 39 |
- config_name: v2.1
|
| 40 |
features:
|
| 41 |
- name: answers
|
|
|
|
| 68 |
num_examples: 101092
|
| 69 |
download_size: 1384271865
|
| 70 |
dataset_size: 4287455242
|
| 71 |
+
configs:
|
| 72 |
+
- config_name: v1.1
|
| 73 |
+
data_files:
|
| 74 |
+
- split: validation
|
| 75 |
+
path: v1.1/validation-*
|
| 76 |
+
- split: train
|
| 77 |
+
path: v1.1/train-*
|
| 78 |
+
- split: test
|
| 79 |
+
path: v1.1/test-*
|
| 80 |
---
|
| 81 |
|
| 82 |
# Dataset Card for "ms_marco"
|
dataset_infos.json
CHANGED
|
@@ -1 +1,199 @@
|
|
| 1 |
-
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"v1.1": {
|
| 3 |
+
"description": "\nStarting with a paper released at NIPS 2016, MS MARCO is a collection of datasets focused on deep learning in search.\n\nThe first dataset was a question answering dataset featuring 100,000 real Bing questions and a human generated answer.\nSince then we released a 1,000,000 question dataset, a natural langauge generation dataset, a passage ranking dataset,\nkeyphrase extraction dataset, crawling dataset, and a conversational search.\n\nThere have been 277 submissions. 20 KeyPhrase Extraction submissions, 87 passage ranking submissions, 0 document ranking\nsubmissions, 73 QnA V2 submissions, 82 NLGEN submisions, and 15 QnA V1 submissions\n\nThis data comes in three tasks/forms: Original QnA dataset(v1.1), Question Answering(v2.1), Natural Language Generation(v2.1).\n\nThe original question answering datset featured 100,000 examples and was released in 2016. Leaderboard is now closed but data is availible below.\n\nThe current competitive tasks are Question Answering and Natural Language Generation. Question Answering features over 1,000,000 queries and\nis much like the original QnA dataset but bigger and with higher quality. The Natural Language Generation dataset features 180,000 examples and\nbuilds upon the QnA dataset to deliver answers that could be spoken by a smart speaker.\n\n\nversion v1.1",
|
| 4 |
+
"citation": "\n@article{DBLP:journals/corr/NguyenRSGTMD16,\n author = {Tri Nguyen and\n Mir Rosenberg and\n Xia Song and\n Jianfeng Gao and\n Saurabh Tiwary and\n Rangan Majumder and\n Li Deng},\n title = {{MS} {MARCO:} {A} Human Generated MAchine Reading COmprehension Dataset},\n journal = {CoRR},\n volume = {abs/1611.09268},\n year = {2016},\n url = {http://arxiv.org/abs/1611.09268},\n archivePrefix = {arXiv},\n eprint = {1611.09268},\n timestamp = {Mon, 13 Aug 2018 16:49:03 +0200},\n biburl = {https://dblp.org/rec/journals/corr/NguyenRSGTMD16.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n}\n}\n",
|
| 5 |
+
"homepage": "https://microsoft.github.io/msmarco/",
|
| 6 |
+
"license": "",
|
| 7 |
+
"features": {
|
| 8 |
+
"answers": {
|
| 9 |
+
"feature": {
|
| 10 |
+
"dtype": "string",
|
| 11 |
+
"_type": "Value"
|
| 12 |
+
},
|
| 13 |
+
"_type": "Sequence"
|
| 14 |
+
},
|
| 15 |
+
"passages": {
|
| 16 |
+
"feature": {
|
| 17 |
+
"is_selected": {
|
| 18 |
+
"dtype": "int32",
|
| 19 |
+
"_type": "Value"
|
| 20 |
+
},
|
| 21 |
+
"passage_text": {
|
| 22 |
+
"dtype": "string",
|
| 23 |
+
"_type": "Value"
|
| 24 |
+
},
|
| 25 |
+
"url": {
|
| 26 |
+
"dtype": "string",
|
| 27 |
+
"_type": "Value"
|
| 28 |
+
}
|
| 29 |
+
},
|
| 30 |
+
"_type": "Sequence"
|
| 31 |
+
},
|
| 32 |
+
"query": {
|
| 33 |
+
"dtype": "string",
|
| 34 |
+
"_type": "Value"
|
| 35 |
+
},
|
| 36 |
+
"query_id": {
|
| 37 |
+
"dtype": "int32",
|
| 38 |
+
"_type": "Value"
|
| 39 |
+
},
|
| 40 |
+
"query_type": {
|
| 41 |
+
"dtype": "string",
|
| 42 |
+
"_type": "Value"
|
| 43 |
+
},
|
| 44 |
+
"wellFormedAnswers": {
|
| 45 |
+
"feature": {
|
| 46 |
+
"dtype": "string",
|
| 47 |
+
"_type": "Value"
|
| 48 |
+
},
|
| 49 |
+
"_type": "Sequence"
|
| 50 |
+
}
|
| 51 |
+
},
|
| 52 |
+
"builder_name": "ms_marco",
|
| 53 |
+
"dataset_name": "ms_marco",
|
| 54 |
+
"config_name": "v1.1",
|
| 55 |
+
"version": {
|
| 56 |
+
"version_str": "1.1.0",
|
| 57 |
+
"description": "",
|
| 58 |
+
"major": 1,
|
| 59 |
+
"minor": 1,
|
| 60 |
+
"patch": 0
|
| 61 |
+
},
|
| 62 |
+
"splits": {
|
| 63 |
+
"validation": {
|
| 64 |
+
"name": "validation",
|
| 65 |
+
"num_bytes": 42665198,
|
| 66 |
+
"num_examples": 10047,
|
| 67 |
+
"dataset_name": null
|
| 68 |
+
},
|
| 69 |
+
"train": {
|
| 70 |
+
"name": "train",
|
| 71 |
+
"num_bytes": 350516260,
|
| 72 |
+
"num_examples": 82326,
|
| 73 |
+
"dataset_name": null
|
| 74 |
+
},
|
| 75 |
+
"test": {
|
| 76 |
+
"name": "test",
|
| 77 |
+
"num_bytes": 40977580,
|
| 78 |
+
"num_examples": 9650,
|
| 79 |
+
"dataset_name": null
|
| 80 |
+
}
|
| 81 |
+
},
|
| 82 |
+
"download_size": 217328153,
|
| 83 |
+
"dataset_size": 434159038,
|
| 84 |
+
"size_in_bytes": 651487191
|
| 85 |
+
},
|
| 86 |
+
"v2.1": {
|
| 87 |
+
"description": "\nStarting with a paper released at NIPS 2016, MS MARCO is a collection of datasets focused on deep learning in search.\n\nThe first dataset was a question answering dataset featuring 100,000 real Bing questions and a human generated answer. \nSince then we released a 1,000,000 question dataset, a natural langauge generation dataset, a passage ranking dataset, \nkeyphrase extraction dataset, crawling dataset, and a conversational search.\n\nThere have been 277 submissions. 20 KeyPhrase Extraction submissions, 87 passage ranking submissions, 0 document ranking \nsubmissions, 73 QnA V2 submissions, 82 NLGEN submisions, and 15 QnA V1 submissions\n\nThis data comes in three tasks/forms: Original QnA dataset(v1.1), Question Answering(v2.1), Natural Language Generation(v2.1). \n\nThe original question answering datset featured 100,000 examples and was released in 2016. Leaderboard is now closed but data is availible below.\n\nThe current competitive tasks are Question Answering and Natural Language Generation. Question Answering features over 1,000,000 queries and \nis much like the original QnA dataset but bigger and with higher quality. The Natural Language Generation dataset features 180,000 examples and \nbuilds upon the QnA dataset to deliver answers that could be spoken by a smart speaker.\n\n\nversion v2.1",
|
| 88 |
+
"citation": "\n@article{DBLP:journals/corr/NguyenRSGTMD16,\n author = {Tri Nguyen and\n Mir Rosenberg and\n Xia Song and\n Jianfeng Gao and\n Saurabh Tiwary and\n Rangan Majumder and\n Li Deng},\n title = {{MS} {MARCO:} {A} Human Generated MAchine Reading COmprehension Dataset},\n journal = {CoRR},\n volume = {abs/1611.09268},\n year = {2016},\n url = {http://arxiv.org/abs/1611.09268},\n archivePrefix = {arXiv},\n eprint = {1611.09268},\n timestamp = {Mon, 13 Aug 2018 16:49:03 +0200},\n biburl = {https://dblp.org/rec/journals/corr/NguyenRSGTMD16.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n}\n}\n",
|
| 89 |
+
"homepage": "https://microsoft.github.io/msmarco/",
|
| 90 |
+
"license": "",
|
| 91 |
+
"features": {
|
| 92 |
+
"answers": {
|
| 93 |
+
"feature": {
|
| 94 |
+
"dtype": "string",
|
| 95 |
+
"id": null,
|
| 96 |
+
"_type": "Value"
|
| 97 |
+
},
|
| 98 |
+
"length": -1,
|
| 99 |
+
"id": null,
|
| 100 |
+
"_type": "Sequence"
|
| 101 |
+
},
|
| 102 |
+
"passages": {
|
| 103 |
+
"feature": {
|
| 104 |
+
"is_selected": {
|
| 105 |
+
"dtype": "int32",
|
| 106 |
+
"id": null,
|
| 107 |
+
"_type": "Value"
|
| 108 |
+
},
|
| 109 |
+
"passage_text": {
|
| 110 |
+
"dtype": "string",
|
| 111 |
+
"id": null,
|
| 112 |
+
"_type": "Value"
|
| 113 |
+
},
|
| 114 |
+
"url": {
|
| 115 |
+
"dtype": "string",
|
| 116 |
+
"id": null,
|
| 117 |
+
"_type": "Value"
|
| 118 |
+
}
|
| 119 |
+
},
|
| 120 |
+
"length": -1,
|
| 121 |
+
"id": null,
|
| 122 |
+
"_type": "Sequence"
|
| 123 |
+
},
|
| 124 |
+
"query": {
|
| 125 |
+
"dtype": "string",
|
| 126 |
+
"id": null,
|
| 127 |
+
"_type": "Value"
|
| 128 |
+
},
|
| 129 |
+
"query_id": {
|
| 130 |
+
"dtype": "int32",
|
| 131 |
+
"id": null,
|
| 132 |
+
"_type": "Value"
|
| 133 |
+
},
|
| 134 |
+
"query_type": {
|
| 135 |
+
"dtype": "string",
|
| 136 |
+
"id": null,
|
| 137 |
+
"_type": "Value"
|
| 138 |
+
},
|
| 139 |
+
"wellFormedAnswers": {
|
| 140 |
+
"feature": {
|
| 141 |
+
"dtype": "string",
|
| 142 |
+
"id": null,
|
| 143 |
+
"_type": "Value"
|
| 144 |
+
},
|
| 145 |
+
"length": -1,
|
| 146 |
+
"id": null,
|
| 147 |
+
"_type": "Sequence"
|
| 148 |
+
}
|
| 149 |
+
},
|
| 150 |
+
"supervised_keys": null,
|
| 151 |
+
"builder_name": "ms_marco",
|
| 152 |
+
"config_name": "v2.1",
|
| 153 |
+
"version": {
|
| 154 |
+
"version_str": "2.1.0",
|
| 155 |
+
"description": "",
|
| 156 |
+
"datasets_version_to_prepare": null,
|
| 157 |
+
"major": 2,
|
| 158 |
+
"minor": 1,
|
| 159 |
+
"patch": 0
|
| 160 |
+
},
|
| 161 |
+
"splits": {
|
| 162 |
+
"validation": {
|
| 163 |
+
"name": "validation",
|
| 164 |
+
"num_bytes": 414286005,
|
| 165 |
+
"num_examples": 101093,
|
| 166 |
+
"dataset_name": "ms_marco"
|
| 167 |
+
},
|
| 168 |
+
"train": {
|
| 169 |
+
"name": "train",
|
| 170 |
+
"num_bytes": 3466972085,
|
| 171 |
+
"num_examples": 808731,
|
| 172 |
+
"dataset_name": "ms_marco"
|
| 173 |
+
},
|
| 174 |
+
"test": {
|
| 175 |
+
"name": "test",
|
| 176 |
+
"num_bytes": 406197152,
|
| 177 |
+
"num_examples": 101092,
|
| 178 |
+
"dataset_name": "ms_marco"
|
| 179 |
+
}
|
| 180 |
+
},
|
| 181 |
+
"download_checksums": {
|
| 182 |
+
"https://msmarco.blob.core.windows.net/msmarco/train_v2.1.json.gz": {
|
| 183 |
+
"num_bytes": 1112116929,
|
| 184 |
+
"checksum": "e91745411ca81e441a3bb75deb71ce000dc2fc31334085b7d499982f14218fe2"
|
| 185 |
+
},
|
| 186 |
+
"https://msmarco.blob.core.windows.net/msmarco/dev_v2.1.json.gz": {
|
| 187 |
+
"num_bytes": 138303699,
|
| 188 |
+
"checksum": "5b3c9c20d1808ee199a930941b0d96f79e397e9234f77a1496890b138df7cb3c"
|
| 189 |
+
},
|
| 190 |
+
"https://msmarco.blob.core.windows.net/msmarco/eval_v2.1_public.json.gz": {
|
| 191 |
+
"num_bytes": 133851237,
|
| 192 |
+
"checksum": "05ac0e448450d507e7ff8e37f48a41cc2d015f5bd2c7974d2445f00a53625db6"
|
| 193 |
+
}
|
| 194 |
+
},
|
| 195 |
+
"download_size": 1384271865,
|
| 196 |
+
"dataset_size": 4287455242,
|
| 197 |
+
"size_in_bytes": 5671727107
|
| 198 |
+
}
|
| 199 |
+
}
|
v1.1/test-00000-of-00001.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28721adaa7767035faa26c647052db537ecc0675ccd85c6d3a559a157bb14f82
|
| 3 |
+
size 20484569
|
v1.1/train-00000-of-00001.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11d5e51d70ce87b47b1c132ef7c55f4cfb93b22c122058aa0591a65c4c0dc1f1
|
| 3 |
+
size 175452226
|
v1.1/validation-00000-of-00001.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67aca2e1b2fffe418a6bd317142887c90973c9c98dd8d940eaae3f8521daed7c
|
| 3 |
+
size 21391358
|