system HF staff commited on
Commit
d3b7dc2
1 Parent(s): 67e339f

Update files from the datasets library (from 1.10.0)

Browse files

Release notes: https://github.com/huggingface/datasets/releases/tag/1.10.0

Files changed (2) hide show
  1. README.md +1 -0
  2. squad.py +8 -10
README.md CHANGED
@@ -1,4 +1,5 @@
1
  ---
 
2
  annotations_creators:
3
  - crowdsourced
4
  language_creators:
1
  ---
2
+ pretty_name: SQuAD
3
  annotations_creators:
4
  - crowdsourced
5
  language_creators:
squad.py CHANGED
@@ -117,28 +117,26 @@ class Squad(datasets.GeneratorBasedBuilder):
117
  def _generate_examples(self, filepath):
118
  """This function returns the examples in the raw (text) form."""
119
  logger.info("generating examples from = %s", filepath)
 
120
  with open(filepath, encoding="utf-8") as f:
121
  squad = json.load(f)
122
  for article in squad["data"]:
123
- title = article.get("title", "").strip()
124
  for paragraph in article["paragraphs"]:
125
- context = paragraph["context"].strip()
126
  for qa in paragraph["qas"]:
127
- question = qa["question"].strip()
128
- id_ = qa["id"]
129
-
130
  answer_starts = [answer["answer_start"] for answer in qa["answers"]]
131
- answers = [answer["text"].strip() for answer in qa["answers"]]
132
-
133
  # Features currently used are "context", "question", and "answers".
134
  # Others are extracted here for the ease of future expansions.
135
- yield id_, {
136
  "title": title,
137
  "context": context,
138
- "question": question,
139
- "id": id_,
140
  "answers": {
141
  "answer_start": answer_starts,
142
  "text": answers,
143
  },
144
  }
 
117
  def _generate_examples(self, filepath):
118
  """This function returns the examples in the raw (text) form."""
119
  logger.info("generating examples from = %s", filepath)
120
+ key = 0
121
  with open(filepath, encoding="utf-8") as f:
122
  squad = json.load(f)
123
  for article in squad["data"]:
124
+ title = article.get("title", "")
125
  for paragraph in article["paragraphs"]:
126
+ context = paragraph["context"] # do not strip leading blank spaces GH-2585
127
  for qa in paragraph["qas"]:
 
 
 
128
  answer_starts = [answer["answer_start"] for answer in qa["answers"]]
129
+ answers = [answer["text"] for answer in qa["answers"]]
 
130
  # Features currently used are "context", "question", and "answers".
131
  # Others are extracted here for the ease of future expansions.
132
+ yield key, {
133
  "title": title,
134
  "context": context,
135
+ "question": qa["question"],
136
+ "id": qa["id"],
137
  "answers": {
138
  "answer_start": answer_starts,
139
  "text": answers,
140
  },
141
  }
142
+ key += 1