Commit
·
98d425e
1
Parent(s):
8e45719
Support streaming (#4)
Browse files- Host data file (4696d9fc2f7f678cd27880718b9f102cefe74496)
- Update and refactor code (46790a05ffbb5a4277937ec95f08970c396398b7)
- data.zip +3 -0
- 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 = "
|
| 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 |
-
|
| 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={
|
|
|
|
|
|
|
|
|
|
| 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": "
|
| 278 |
},
|
| 279 |
),
|
| 280 |
]
|
| 281 |
|
| 282 |
-
def _generate_examples(self,
|
| 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 |
|