Update SimCLUE.py
Browse files- SimCLUE.py +9 -2
SimCLUE.py
CHANGED
@@ -40,6 +40,13 @@ class SimCLUE(datasets.GeneratorBasedBuilder):
|
|
40 |
'dev': 'dev.json',
|
41 |
'test_public': 'test_public.json'}
|
42 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
43 |
BUILDER_CONFIGS = [
|
44 |
SimCLUEConfig(
|
45 |
name="train_rank",
|
@@ -112,7 +119,7 @@ class SimCLUE(datasets.GeneratorBasedBuilder):
|
|
112 |
def _split_generators(self, dl_manager):
|
113 |
dl_dir = dl_manager.download_and_extract(self.config.data_url)
|
114 |
return [datasets.SplitGenerator(
|
115 |
-
name=self.config.name,
|
116 |
gen_kwargs={
|
117 |
"filepath": os.path.join(dl_dir, self.part_file[self.config.name]),
|
118 |
})]
|
@@ -121,4 +128,4 @@ class SimCLUE(datasets.GeneratorBasedBuilder):
|
|
121 |
"""This function returns the examples in the raw (text) form."""
|
122 |
with open(filepath, 'r', encoding="utf-8") as f:
|
123 |
for idx, row in enumerate(f):
|
124 |
-
yield idx, json.loads(row)
|
|
|
40 |
'dev': 'dev.json',
|
41 |
'test_public': 'test_public.json'}
|
42 |
|
43 |
+
part_split = {'train_rank': datasets.Split.TRAIN,
|
44 |
+
'train_pair': datasets.Split.TRAIN,
|
45 |
+
'corpus': datasets.Split.TRAIN,
|
46 |
+
'train_pair_postive': datasets.Split.TRAIN,
|
47 |
+
'dev': datasets.Split.VALIDATION,
|
48 |
+
'test_public': datasets.Split.TEST}
|
49 |
+
|
50 |
BUILDER_CONFIGS = [
|
51 |
SimCLUEConfig(
|
52 |
name="train_rank",
|
|
|
119 |
def _split_generators(self, dl_manager):
|
120 |
dl_dir = dl_manager.download_and_extract(self.config.data_url)
|
121 |
return [datasets.SplitGenerator(
|
122 |
+
name=self.part_split[self.config.name],
|
123 |
gen_kwargs={
|
124 |
"filepath": os.path.join(dl_dir, self.part_file[self.config.name]),
|
125 |
})]
|
|
|
128 |
"""This function returns the examples in the raw (text) form."""
|
129 |
with open(filepath, 'r', encoding="utf-8") as f:
|
130 |
for idx, row in enumerate(f):
|
131 |
+
yield idx, json.loads(row)
|