qanastek commited on
Commit
f1f0a39
1 Parent(s): f899be3
Files changed (1) hide show
  1. ELRC-Medical-V2.py +27 -11
ELRC-Medical-V2.py CHANGED
@@ -56,15 +56,21 @@ class ELRC_Medical_V2(datasets.GeneratorBasedBuilder):
56
  DEFAULT_CONFIG_NAME = "en-fr"
57
 
58
  def _info(self):
 
 
59
  return datasets.DatasetInfo(
60
  description=_DESCRIPTION,
61
- features=datasets.Features({
62
- "doc_id": datasets.Value("int32"),
63
- "lang": datasets.Value("string"),
64
- "source_text": datasets.Value("string"),
65
- "target_text": datasets.Value("string"),
66
- }),
67
- supervised_keys=None,
 
 
 
 
68
  homepage="https://github.com/qanastek/ELRC-Medical-V2/",
69
  citation=_CITATION,
70
  license=_LICENSE,
@@ -91,6 +97,9 @@ class ELRC_Medical_V2(datasets.GeneratorBasedBuilder):
91
 
92
  logger.info("⏳ Generating examples from = %s", filepath)
93
 
 
 
 
94
  key_ = 0
95
 
96
  with open(filepath, encoding="utf-8") as f:
@@ -101,10 +110,17 @@ class ELRC_Medical_V2(datasets.GeneratorBasedBuilder):
101
  continue
102
 
103
  yield key_, {
104
- "doc_id": int(row[0]),
105
- "lang": str(row[1]),
106
- "source_text": str(row[2]),
107
- "target_text": str(row[3])
108
  }
109
 
 
 
 
 
 
 
 
110
  key_ += 1
 
56
  DEFAULT_CONFIG_NAME = "en-fr"
57
 
58
  def _info(self):
59
+ src, target = self.config.name.split("-")
60
+ pair = (src, target)
61
  return datasets.DatasetInfo(
62
  description=_DESCRIPTION,
63
+ features=datasets.Features(
64
+ {"translation": datasets.features.Translation(languages=pair)}
65
+ ),
66
+ supervised_keys=(src, target),
67
+ # features=datasets.Features({
68
+ # "doc_id": datasets.Value("int32"),
69
+ # "lang": datasets.Value("string"),
70
+ # "source_text": datasets.Value("string"),
71
+ # "target_text": datasets.Value("string"),
72
+ # }),
73
+ # supervised_keys=None,
74
  homepage="https://github.com/qanastek/ELRC-Medical-V2/",
75
  citation=_CITATION,
76
  license=_LICENSE,
 
97
 
98
  logger.info("⏳ Generating examples from = %s", filepath)
99
 
100
+ src, target = str(row[1]).split("-")
101
+ pair = (src, target)
102
+
103
  key_ = 0
104
 
105
  with open(filepath, encoding="utf-8") as f:
 
110
  continue
111
 
112
  yield key_, {
113
+ "translation": {
114
+ src: str(row[2]).strip(),
115
+ target: str(row[3]).strip(),
116
+ }
117
  }
118
 
119
+ # yield key_, {
120
+ # "doc_id": int(row[0]),
121
+ # "lang": str(row[1]),
122
+ # "source_text": str(row[2]),
123
+ # "target_text": str(row[3])
124
+ # }
125
+
126
  key_ += 1