tianbaoxiexxx commited on
Commit
fade991
1 Parent(s): c691da3

Fix bugs in msr_sqa dataset (#3715)

Browse files

* Fix problems in msr_sqa

* Update metadata JSON

* Update version

* Update dummy data version

* Update metadata JSON

Co-authored-by: Tianbao Xie <[email protected]>
Co-authored-by: Albert Villanova del Moral <[email protected]>

Commit from https://github.com/huggingface/datasets/commit/55924c5e3b823a3b1206269bb0892cd3a9508570

dataset_infos.json CHANGED
@@ -1 +1 @@
1
- {"default": {"description": "Recent work in semantic parsing for question answering has focused on long and complicated questions, many of which would seem unnatural if asked in a normal conversation between two humans. In an effort to explore a conversational QA setting, we present a more realistic task: answering sequences of simple but inter-related questions. We created SQA by asking crowdsourced workers to decompose 2,022 questions from WikiTableQuestions (WTQ), which contains highly-compositional questions about tables from Wikipedia. We had three workers decompose each WTQ question, resulting in a dataset of 6,066 sequences that contain 17,553 questions in total. Each question is also associated with answers in the form of cell locations in the tables.\n", "citation": "@inproceedings{iyyer2017search,\n title={Search-based neural structured learning for sequential question answering},\n author={Iyyer, Mohit and Yih, Wen-tau and Chang, Ming-Wei},\n booktitle={Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},\n pages={1821--1831},\n year={2017}\n}\n", "homepage": "https://msropendata.com/datasets/b25190ed-0f59-47b1-9211-5962858142c2", "license": "Microsoft Research Data License Agreement", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "annotator": {"dtype": "int32", "id": null, "_type": "Value"}, "position": {"dtype": "int32", "id": null, "_type": "Value"}, "question": {"dtype": "string", "id": null, "_type": "Value"}, "table_file": {"dtype": "string", "id": null, "_type": "Value"}, "table_header": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "table_data": {"feature": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "length": -1, "id": null, "_type": "Sequence"}, "answer_coordinates": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "answer_text": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": null, "builder_name": "msr_sqa", "config_name": "default", "version": {"version_str": "0.0.0", "description": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 22605449, "num_examples": 14541, "dataset_name": "msr_sqa"}, "test": {"name": "test", "num_bytes": 4924516, "num_examples": 3012, "dataset_name": "msr_sqa"}}, "download_checksums": {"https://download.microsoft.com/download/1/D/C/1DC270D2-1B53-4A61-A2E3-88AB3E4E6E1F/SQA%20Release%201.0.zip": {"num_bytes": 4796932, "checksum": "791a07ef90d6e736c186b25009d3c10cb38624b879bb668033445a3ab8892f64"}}, "download_size": 4796932, "post_processing_size": null, "dataset_size": 27529965, "size_in_bytes": 32326897}}
 
1
+ {"default": {"description": "Recent work in semantic parsing for question answering has focused on long and complicated questions, many of which would seem unnatural if asked in a normal conversation between two humans. In an effort to explore a conversational QA setting, we present a more realistic task: answering sequences of simple but inter-related questions. We created SQA by asking crowdsourced workers to decompose 2,022 questions from WikiTableQuestions (WTQ), which contains highly-compositional questions about tables from Wikipedia. We had three workers decompose each WTQ question, resulting in a dataset of 6,066 sequences that contain 17,553 questions in total. Each question is also associated with answers in the form of cell locations in the tables.\n", "citation": "@inproceedings{iyyer2017search,\n title={Search-based neural structured learning for sequential question answering},\n author={Iyyer, Mohit and Yih, Wen-tau and Chang, Ming-Wei},\n booktitle={Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},\n pages={1821--1831},\n year={2017}\n}\n", "homepage": "https://msropendata.com/datasets/b25190ed-0f59-47b1-9211-5962858142c2", "license": "Microsoft Research Data License Agreement", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "annotator": {"dtype": "int32", "id": null, "_type": "Value"}, "position": {"dtype": "int32", "id": null, "_type": "Value"}, "question": {"dtype": "string", "id": null, "_type": "Value"}, "question_and_history": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "table_file": {"dtype": "string", "id": null, "_type": "Value"}, "table_header": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "table_data": {"feature": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "length": -1, "id": null, "_type": "Sequence"}, "answer_coordinates": {"feature": {"row_index": {"dtype": "int32", "id": null, "_type": "Value"}, "column_index": {"dtype": "int32", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "answer_text": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "msr_sqa", "config_name": "default", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 19732499, "num_examples": 12276, "dataset_name": "msr_sqa"}, "validation": {"name": "validation", "num_bytes": 3738331, "num_examples": 2265, "dataset_name": "msr_sqa"}, "test": {"name": "test", "num_bytes": 5105873, "num_examples": 3012, "dataset_name": "msr_sqa"}}, "download_checksums": {"https://download.microsoft.com/download/1/D/C/1DC270D2-1B53-4A61-A2E3-88AB3E4E6E1F/SQA%20Release%201.0.zip": {"num_bytes": 4796932, "checksum": "791a07ef90d6e736c186b25009d3c10cb38624b879bb668033445a3ab8892f64"}}, "download_size": 4796932, "post_processing_size": null, "dataset_size": 28576703, "size_in_bytes": 33373635}}
dummy/{0.0.0 → 1.0.0}/dummy_data.zip RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7e46d5f939a1049a45c605ba21355084b0043e84b5dc6c7dec2717e0aa326510
3
- size 2732
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81f4cf20ef83ae9682c85ccd9e54aafd0c10dc785980f3a844d0c59b120bc525
3
+ size 445705
msr_sqa.py CHANGED
@@ -19,10 +19,11 @@ import ast
19
  import csv
20
  import os
21
 
 
 
22
  import datasets
23
 
24
 
25
- # TODO: Add BibTeX citation
26
  # Find for instance the citation on arxiv or on the dataset repo/website
27
  _CITATION = """\
28
  @inproceedings{iyyer2017search,
@@ -60,13 +61,16 @@ def _load_table_data(table_file):
60
 
61
  Returns:
62
  header: a list of headers in the table.
63
- data: 2d array of data in the table.
64
  """
65
- with open(table_file, encoding="utf-8") as f:
66
- lines = f.readlines()
67
- header = lines[0].strip().split(",")
68
- data = [line.strip().split(",") for line in lines[1:]]
69
- return header, data
 
 
 
70
 
71
 
72
  def _parse_answer_coordinates(answer_coordinate_str):
@@ -113,6 +117,8 @@ def _parse_answer_text(answer_text_str):
113
  class MsrSQA(datasets.GeneratorBasedBuilder):
114
  """Microsoft Research Sequential Question Answering (SQA) Dataset"""
115
 
 
 
116
  def _info(self):
117
  return datasets.DatasetInfo(
118
  description=_DESCRIPTION,
@@ -122,6 +128,7 @@ class MsrSQA(datasets.GeneratorBasedBuilder):
122
  "annotator": datasets.Value("int32"),
123
  "position": datasets.Value("int32"),
124
  "question": datasets.Value("string"),
 
125
  "table_file": datasets.Value("string"),
126
  "table_header": datasets.features.Sequence(datasets.Value("string")),
127
  "table_data": datasets.features.Sequence(datasets.features.Sequence(datasets.Value("string"))),
@@ -143,7 +150,11 @@ class MsrSQA(datasets.GeneratorBasedBuilder):
143
  return [
144
  datasets.SplitGenerator(
145
  name=datasets.Split.TRAIN,
146
- gen_kwargs={"filepath": os.path.join(data_dir, "train.tsv"), "data_dir": data_dir},
 
 
 
 
147
  ),
148
  datasets.SplitGenerator(
149
  name=datasets.Split.TEST,
@@ -155,10 +166,15 @@ class MsrSQA(datasets.GeneratorBasedBuilder):
155
  """Yields examples."""
156
  with open(filepath, encoding="utf-8") as f:
157
  reader = csv.DictReader(f, delimiter="\t")
 
158
  for idx, item in enumerate(reader):
159
  item["answer_text"] = _parse_answer_text(item["answer_text"])
160
  item["answer_coordinates"] = _parse_answer_coordinates(item["answer_coordinates"])
161
  header, table_data = _load_table_data(os.path.join(data_dir, item["table_file"]))
162
  item["table_header"] = header
163
  item["table_data"] = table_data
 
 
 
 
164
  yield idx, item
 
19
  import csv
20
  import os
21
 
22
+ import pandas as pd
23
+
24
  import datasets
25
 
26
 
 
27
  # Find for instance the citation on arxiv or on the dataset repo/website
28
  _CITATION = """\
29
  @inproceedings{iyyer2017search,
 
61
 
62
  Returns:
63
  header: a list of headers in the table.
64
+ rows: 2d array of data in the table.
65
  """
66
+ rows = []
67
+ table_data = pd.read_csv(table_file)
68
+ # the first line is header
69
+ header = list(table_data.columns)
70
+ for row_data in table_data.values:
71
+ rows.append([str(_) for _ in list(row_data)])
72
+
73
+ return header, rows
74
 
75
 
76
  def _parse_answer_coordinates(answer_coordinate_str):
 
117
  class MsrSQA(datasets.GeneratorBasedBuilder):
118
  """Microsoft Research Sequential Question Answering (SQA) Dataset"""
119
 
120
+ VERSION = datasets.Version("1.0.0")
121
+
122
  def _info(self):
123
  return datasets.DatasetInfo(
124
  description=_DESCRIPTION,
 
128
  "annotator": datasets.Value("int32"),
129
  "position": datasets.Value("int32"),
130
  "question": datasets.Value("string"),
131
+ "question_and_history": datasets.Sequence(datasets.Value("string")),
132
  "table_file": datasets.Value("string"),
133
  "table_header": datasets.features.Sequence(datasets.Value("string")),
134
  "table_data": datasets.features.Sequence(datasets.features.Sequence(datasets.Value("string"))),
 
150
  return [
151
  datasets.SplitGenerator(
152
  name=datasets.Split.TRAIN,
153
+ gen_kwargs={"filepath": os.path.join(data_dir, "random-split-1-train.tsv"), "data_dir": data_dir},
154
+ ),
155
+ datasets.SplitGenerator(
156
+ name=datasets.Split.VALIDATION,
157
+ gen_kwargs={"filepath": os.path.join(data_dir, "random-split-1-dev.tsv"), "data_dir": data_dir},
158
  ),
159
  datasets.SplitGenerator(
160
  name=datasets.Split.TEST,
 
166
  """Yields examples."""
167
  with open(filepath, encoding="utf-8") as f:
168
  reader = csv.DictReader(f, delimiter="\t")
169
+ question_and_history = []
170
  for idx, item in enumerate(reader):
171
  item["answer_text"] = _parse_answer_text(item["answer_text"])
172
  item["answer_coordinates"] = _parse_answer_coordinates(item["answer_coordinates"])
173
  header, table_data = _load_table_data(os.path.join(data_dir, item["table_file"]))
174
  item["table_header"] = header
175
  item["table_data"] = table_data
176
+ if item["position"] == "0":
177
+ question_and_history = [] # reset history
178
+ question_and_history.append(item["question"])
179
+ item["question_and_history"] = question_and_history
180
  yield idx, item