Update DEFT2023.py
Browse files- DEFT2023.py +7 -3
DEFT2023.py
CHANGED
@@ -27,6 +27,8 @@ _HOMEPAGE = "https://frenchmedmcqa.github.io"
|
|
27 |
|
28 |
_LICENSE = "Apache License 2.0"
|
29 |
|
|
|
|
|
30 |
_CITATION = """\
|
31 |
@unpublished{labrak:hal-03824241,
|
32 |
TITLE = {{FrenchMedMCQA: A French Multiple-Choice Question Answering Dataset for Medical domain}},
|
@@ -77,23 +79,25 @@ class DEFT2023(datasets.GeneratorBasedBuilder):
|
|
77 |
def _split_generators(self, dl_manager):
|
78 |
"""Returns SplitGenerators."""
|
79 |
|
|
|
|
|
80 |
return [
|
81 |
datasets.SplitGenerator(
|
82 |
name=datasets.Split.TRAIN,
|
83 |
gen_kwargs={
|
84 |
-
"filepath": os.path.join(
|
85 |
},
|
86 |
),
|
87 |
datasets.SplitGenerator(
|
88 |
name=datasets.Split.VALIDATION,
|
89 |
gen_kwargs={
|
90 |
-
"filepath": os.path.join(
|
91 |
},
|
92 |
),
|
93 |
datasets.SplitGenerator(
|
94 |
name=datasets.Split.TEST,
|
95 |
gen_kwargs={
|
96 |
-
"filepath": os.path.join(
|
97 |
},
|
98 |
),
|
99 |
]
|
|
|
27 |
|
28 |
_LICENSE = "Apache License 2.0"
|
29 |
|
30 |
+
_URL = "https://huggingface.co/datasets/Dr-BERT/FrenchMedMCQA/resolve/main/data.zip"
|
31 |
+
|
32 |
_CITATION = """\
|
33 |
@unpublished{labrak:hal-03824241,
|
34 |
TITLE = {{FrenchMedMCQA: A French Multiple-Choice Question Answering Dataset for Medical domain}},
|
|
|
79 |
def _split_generators(self, dl_manager):
|
80 |
"""Returns SplitGenerators."""
|
81 |
|
82 |
+
data_dir = dl_manager.download_and_extract(_URL)
|
83 |
+
|
84 |
return [
|
85 |
datasets.SplitGenerator(
|
86 |
name=datasets.Split.TRAIN,
|
87 |
gen_kwargs={
|
88 |
+
"filepath": os.path.join(data_dir, "/data/train.json"),
|
89 |
},
|
90 |
),
|
91 |
datasets.SplitGenerator(
|
92 |
name=datasets.Split.VALIDATION,
|
93 |
gen_kwargs={
|
94 |
+
"filepath": os.path.join(data_dir, "/data/dev.json"),
|
95 |
},
|
96 |
),
|
97 |
datasets.SplitGenerator(
|
98 |
name=datasets.Split.TEST,
|
99 |
gen_kwargs={
|
100 |
+
"filepath": os.path.join(data_dir, "/data/test.json"),
|
101 |
},
|
102 |
),
|
103 |
]
|