albertvillanova HF staff commited on
Commit
98d425e
1 Parent(s): 8e45719

Support streaming (#4)

Browse files

- Host data file (4696d9fc2f7f678cd27880718b9f102cefe74496)
- Update and refactor code (46790a05ffbb5a4277937ec95f08970c396398b7)

Files changed (2) hide show
  1. data.zip +3 -0
  2. sofc_materials_articles.py +12 -28
data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5430c7563887db6067e75bcfbad99dfa83b02b38834dbf5cfd355919d84554ec
3
+ size 3627968
sofc_materials_articles.py CHANGED
@@ -45,7 +45,7 @@ _HOMEPAGE = "https://arxiv.org/abs/2006.03039"
45
 
46
  _LICENSE = ""
47
 
48
- _URL = "https://github.com/boschresearch/sofc-exp_textmining_resources/archive/master.zip"
49
 
50
 
51
  class SOFCMaterialsArticles(datasets.GeneratorBasedBuilder):
@@ -232,33 +232,14 @@ class SOFCMaterialsArticles(datasets.GeneratorBasedBuilder):
232
 
233
  def _split_generators(self, dl_manager):
234
  """Returns SplitGenerators."""
235
-
236
- my_urls = _URL
237
- data_dir = dl_manager.download_and_extract(my_urls)
238
-
239
- data_dir = os.path.join(data_dir, "sofc-exp_textmining_resources-master/sofc-exp-corpus")
240
-
241
- metadata = pd.read_csv(os.path.join(data_dir, "SOFC-Exp-Metadata.csv"), sep="\t")
242
-
243
- text_base_path = os.path.join(data_dir, "texts")
244
-
245
- text_files_available = [
246
- os.path.split(i.rstrip(".txt"))[-1] for i in glob.glob(os.path.join(text_base_path, "*.txt"))
247
- ]
248
-
249
- metadata = metadata[metadata["name"].map(lambda x: x in text_files_available)]
250
-
251
- names = {}
252
- splits = ["train", "test", "dev"]
253
- for split in splits:
254
- names[split] = metadata[metadata["set"] == split]["name"].tolist()
255
 
256
  return [
257
  datasets.SplitGenerator(
258
  name=datasets.Split.TRAIN,
259
  # These kwargs will be passed to _generate_examples
260
  gen_kwargs={
261
- "names": names["train"],
262
  "data_dir": data_dir,
263
  "split": "train",
264
  },
@@ -266,21 +247,26 @@ class SOFCMaterialsArticles(datasets.GeneratorBasedBuilder):
266
  datasets.SplitGenerator(
267
  name=datasets.Split.TEST,
268
  # These kwargs will be passed to _generate_examples
269
- gen_kwargs={"names": names["test"], "data_dir": data_dir, "split": "test"},
 
 
 
270
  ),
271
  datasets.SplitGenerator(
272
  name=datasets.Split.VALIDATION,
273
  # These kwargs will be passed to _generate_examples
274
  gen_kwargs={
275
- "names": names["dev"],
276
  "data_dir": data_dir,
277
- "split": "validation",
278
  },
279
  ),
280
  ]
281
 
282
- def _generate_examples(self, names, data_dir, split):
283
  """Yields examples."""
 
 
 
284
  # The dataset consists of the original article text as well as annotations
285
  textfile_base_path = os.path.join(data_dir, "texts")
286
  annotations_base_path = os.path.join(data_dir, "annotations")
@@ -308,7 +294,6 @@ class SOFCMaterialsArticles(datasets.GeneratorBasedBuilder):
308
  # For each text file, we'll load all of the
309
  # associated annotation files
310
  for id_, name in enumerate(sorted(names)):
311
-
312
  # Load the main source text
313
  textfile_path = os.path.join(textfile_base_path, name + ".txt")
314
  text = open(textfile_path, encoding="utf-8").read()
@@ -383,7 +368,6 @@ class SOFCMaterialsArticles(datasets.GeneratorBasedBuilder):
383
  # Iterate through the spans data
384
  spans = []
385
  for span in spans_raw:
386
-
387
  # Split out the elements in each tab-delimited line
388
  _, span_id, entity_label_or_exp, sentence_id, begin_char_offset, end_char_offset = span.split("\t")
389
 
45
 
46
  _LICENSE = ""
47
 
48
+ _URL = "data.zip"
49
 
50
 
51
  class SOFCMaterialsArticles(datasets.GeneratorBasedBuilder):
232
 
233
  def _split_generators(self, dl_manager):
234
  """Returns SplitGenerators."""
235
+ data_dir = dl_manager.download_and_extract(_URL)
236
+ data_dir = os.path.join(data_dir, "sofc-exp-corpus")
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
237
 
238
  return [
239
  datasets.SplitGenerator(
240
  name=datasets.Split.TRAIN,
241
  # These kwargs will be passed to _generate_examples
242
  gen_kwargs={
 
243
  "data_dir": data_dir,
244
  "split": "train",
245
  },
247
  datasets.SplitGenerator(
248
  name=datasets.Split.TEST,
249
  # These kwargs will be passed to _generate_examples
250
+ gen_kwargs={
251
+ "data_dir": data_dir,
252
+ "split": "test",
253
+ },
254
  ),
255
  datasets.SplitGenerator(
256
  name=datasets.Split.VALIDATION,
257
  # These kwargs will be passed to _generate_examples
258
  gen_kwargs={
 
259
  "data_dir": data_dir,
260
+ "split": "dev",
261
  },
262
  ),
263
  ]
264
 
265
+ def _generate_examples(self, data_dir, split):
266
  """Yields examples."""
267
+ metadata = pd.read_csv(os.path.join(data_dir, "SOFC-Exp-Metadata.csv"), sep="\t")
268
+ names = metadata[metadata["set"] == split]["name"].tolist()
269
+
270
  # The dataset consists of the original article text as well as annotations
271
  textfile_base_path = os.path.join(data_dir, "texts")
272
  annotations_base_path = os.path.join(data_dir, "annotations")
294
  # For each text file, we'll load all of the
295
  # associated annotation files
296
  for id_, name in enumerate(sorted(names)):
 
297
  # Load the main source text
298
  textfile_path = os.path.join(textfile_base_path, name + ".txt")
299
  text = open(textfile_path, encoding="utf-8").read()
368
  # Iterate through the spans data
369
  spans = []
370
  for span in spans_raw:
 
371
  # Split out the elements in each tab-delimited line
372
  _, span_id, entity_label_or_exp, sentence_id, begin_char_offset, end_char_offset = span.split("\t")
373