Update scripts to use data.zip instead of local data.
Browse files- DEFT2020.py +24 -21
- data.zip +2 -2
- test_dataset.py +0 -9
DEFT2020.py
CHANGED
@@ -62,9 +62,14 @@ _CITATION = """\
|
|
62 |
}
|
63 |
"""
|
64 |
|
65 |
-
|
|
|
|
|
|
|
66 |
"""DEFT 2020 - DEFT (DÉfi Fouille de Textes)"""
|
67 |
|
|
|
|
|
68 |
VERSION = datasets.Version("1.0.0")
|
69 |
|
70 |
BUILDER_CONFIGS = [
|
@@ -72,15 +77,13 @@ class FrenchMedMCQA(datasets.GeneratorBasedBuilder):
|
|
72 |
datasets.BuilderConfig(name="task_2", version=VERSION, description="Tâche 2 : identifier les phrases parallèles possible pour une phrase source"),
|
73 |
]
|
74 |
|
75 |
-
DEFAULT_CONFIG_NAME = "task_1"
|
76 |
-
|
77 |
def _info(self):
|
78 |
|
79 |
if self.config.name == "default":
|
80 |
self.config.name = self.DEFAULT_CONFIG_NAME
|
81 |
|
82 |
if self.config.name == "task_1":
|
83 |
-
|
84 |
return datasets.DatasetInfo(
|
85 |
description=_DESCRIPTION,
|
86 |
features=datasets.Features(
|
@@ -98,16 +101,16 @@ class FrenchMedMCQA(datasets.GeneratorBasedBuilder):
|
|
98 |
citation=_CITATION,
|
99 |
license=_LICENSE,
|
100 |
)
|
101 |
-
|
102 |
elif self.config.name == "task_2":
|
103 |
-
|
104 |
return datasets.DatasetInfo(
|
105 |
description=_DESCRIPTION,
|
106 |
features=datasets.Features(
|
107 |
{
|
108 |
"id": datasets.Value("string"),
|
109 |
"correct_cible": datasets.features.ClassLabel(
|
110 |
-
names
|
111 |
),
|
112 |
"source": datasets.Value("string"),
|
113 |
"cible_1": datasets.Value("string"),
|
@@ -124,7 +127,7 @@ class FrenchMedMCQA(datasets.GeneratorBasedBuilder):
|
|
124 |
def _split_generators(self, dl_manager):
|
125 |
"""Returns SplitGenerators."""
|
126 |
|
127 |
-
data_dir =
|
128 |
|
129 |
if self.config.name == "task_1":
|
130 |
|
@@ -183,16 +186,16 @@ class FrenchMedMCQA(datasets.GeneratorBasedBuilder):
|
|
183 |
all_res = []
|
184 |
|
185 |
if self.config.name == "task_1":
|
186 |
-
|
187 |
ratio = 0.8316
|
188 |
-
|
189 |
with open(filepath) as fd:
|
190 |
doc = xmltodict.parse(fd.read())
|
191 |
|
192 |
for d in doc["doc"]["paire"]:
|
193 |
|
194 |
if int(d["@id"]) > -1:
|
195 |
-
|
196 |
all_res.append({
|
197 |
"id": d["@id"],
|
198 |
"vote": d["@vote"],
|
@@ -203,9 +206,9 @@ class FrenchMedMCQA(datasets.GeneratorBasedBuilder):
|
|
203 |
})
|
204 |
|
205 |
elif self.config.name == "task_2":
|
206 |
-
|
207 |
ratio = 0.8059
|
208 |
-
|
209 |
with open(filepath) as fd:
|
210 |
doc = xmltodict.parse(fd.read())
|
211 |
|
@@ -222,29 +225,29 @@ class FrenchMedMCQA(datasets.GeneratorBasedBuilder):
|
|
222 |
|
223 |
for t in d["cible"]:
|
224 |
obj[f"cible_{t['@num']}"] = t["#text"]
|
225 |
-
|
226 |
all_res.append(obj)
|
227 |
|
228 |
if split != "test":
|
229 |
-
|
230 |
ids = [r["id"] for r in all_res]
|
231 |
-
|
232 |
random.seed(4)
|
233 |
random.shuffle(ids)
|
234 |
random.shuffle(ids)
|
235 |
random.shuffle(ids)
|
236 |
-
|
237 |
-
train, validation = np.split(ids, [int(len(ids)*ratio)])
|
238 |
-
|
239 |
if split == "train":
|
240 |
allowed_ids = list(train)
|
241 |
elif split == "validation":
|
242 |
allowed_ids = list(validation)
|
243 |
-
|
244 |
for r in all_res:
|
245 |
if r["id"] in allowed_ids:
|
246 |
yield r["id"], r
|
247 |
else:
|
248 |
-
|
249 |
for r in all_res:
|
250 |
yield r["id"], r
|
|
|
62 |
}
|
63 |
"""
|
64 |
|
65 |
+
_URL = "data.zip"
|
66 |
+
|
67 |
+
|
68 |
+
class DEFT2020(datasets.GeneratorBasedBuilder):
|
69 |
"""DEFT 2020 - DEFT (DÉfi Fouille de Textes)"""
|
70 |
|
71 |
+
DEFAULT_CONFIG_NAME = "task_1"
|
72 |
+
|
73 |
VERSION = datasets.Version("1.0.0")
|
74 |
|
75 |
BUILDER_CONFIGS = [
|
|
|
77 |
datasets.BuilderConfig(name="task_2", version=VERSION, description="Tâche 2 : identifier les phrases parallèles possible pour une phrase source"),
|
78 |
]
|
79 |
|
|
|
|
|
80 |
def _info(self):
|
81 |
|
82 |
if self.config.name == "default":
|
83 |
self.config.name = self.DEFAULT_CONFIG_NAME
|
84 |
|
85 |
if self.config.name == "task_1":
|
86 |
+
|
87 |
return datasets.DatasetInfo(
|
88 |
description=_DESCRIPTION,
|
89 |
features=datasets.Features(
|
|
|
101 |
citation=_CITATION,
|
102 |
license=_LICENSE,
|
103 |
)
|
104 |
+
|
105 |
elif self.config.name == "task_2":
|
106 |
+
|
107 |
return datasets.DatasetInfo(
|
108 |
description=_DESCRIPTION,
|
109 |
features=datasets.Features(
|
110 |
{
|
111 |
"id": datasets.Value("string"),
|
112 |
"correct_cible": datasets.features.ClassLabel(
|
113 |
+
names=['1', '2', '3'],
|
114 |
),
|
115 |
"source": datasets.Value("string"),
|
116 |
"cible_1": datasets.Value("string"),
|
|
|
127 |
def _split_generators(self, dl_manager):
|
128 |
"""Returns SplitGenerators."""
|
129 |
|
130 |
+
data_dir = dl_manager.download_and_extract(_URL).rstrip("/")
|
131 |
|
132 |
if self.config.name == "task_1":
|
133 |
|
|
|
186 |
all_res = []
|
187 |
|
188 |
if self.config.name == "task_1":
|
189 |
+
|
190 |
ratio = 0.8316
|
191 |
+
|
192 |
with open(filepath) as fd:
|
193 |
doc = xmltodict.parse(fd.read())
|
194 |
|
195 |
for d in doc["doc"]["paire"]:
|
196 |
|
197 |
if int(d["@id"]) > -1:
|
198 |
+
|
199 |
all_res.append({
|
200 |
"id": d["@id"],
|
201 |
"vote": d["@vote"],
|
|
|
206 |
})
|
207 |
|
208 |
elif self.config.name == "task_2":
|
209 |
+
|
210 |
ratio = 0.8059
|
211 |
+
|
212 |
with open(filepath) as fd:
|
213 |
doc = xmltodict.parse(fd.read())
|
214 |
|
|
|
225 |
|
226 |
for t in d["cible"]:
|
227 |
obj[f"cible_{t['@num']}"] = t["#text"]
|
228 |
+
|
229 |
all_res.append(obj)
|
230 |
|
231 |
if split != "test":
|
232 |
+
|
233 |
ids = [r["id"] for r in all_res]
|
234 |
+
|
235 |
random.seed(4)
|
236 |
random.shuffle(ids)
|
237 |
random.shuffle(ids)
|
238 |
random.shuffle(ids)
|
239 |
+
|
240 |
+
train, validation = np.split(ids, [int(len(ids) * ratio)])
|
241 |
+
|
242 |
if split == "train":
|
243 |
allowed_ids = list(train)
|
244 |
elif split == "validation":
|
245 |
allowed_ids = list(validation)
|
246 |
+
|
247 |
for r in all_res:
|
248 |
if r["id"] in allowed_ids:
|
249 |
yield r["id"], r
|
250 |
else:
|
251 |
+
|
252 |
for r in all_res:
|
253 |
yield r["id"], r
|
data.zip
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d4d69417b3e74595ae78ade05a772f338840af016e073a5d75e1042e2b77cb3a
|
3 |
+
size 300044
|
test_dataset.py
DELETED
@@ -1,9 +0,0 @@
|
|
1 |
-
from datasets import load_dataset
|
2 |
-
|
3 |
-
dataset = load_dataset("./DEFT2020.py", data_dir="./DEFT2020_T1-T2/", name="task_1")
|
4 |
-
print(dataset)
|
5 |
-
print(dataset["test"][0])
|
6 |
-
|
7 |
-
dataset = load_dataset("./DEFT2020.py", data_dir="./DEFT2020_T1-T2/", name="task_2")
|
8 |
-
print(dataset)
|
9 |
-
print(dataset["test"][0])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|