vipulraheja
commited on
Commit
•
c3576ab
1
Parent(s):
06ba0d0
fix: fix issues in dataloader
Browse files- simpitiki.py +7 -8
simpitiki.py
CHANGED
@@ -61,8 +61,6 @@ _URLs = {
|
|
61 |
"itwiki_test":"./v1/source_dataset_split/itwiki_test.json",
|
62 |
"tn_test":"./v1/source_dataset_split/tn_test.json"
|
63 |
}
|
64 |
-
|
65 |
-
|
66 |
},
|
67 |
"v2":{
|
68 |
"random": {
|
@@ -117,6 +115,7 @@ class SIMPITIKI(datasets.GeneratorBasedBuilder):
|
|
117 |
# This method specifies the datasets.DatasetInfo object which contains informations and typings for the dataset
|
118 |
features = datasets.Features(
|
119 |
{
|
|
|
120 |
"text": datasets.Value("string"),
|
121 |
"simplified_text": datasets.Value("string"),
|
122 |
"transformation_type":datasets.Value("string"),
|
@@ -201,7 +200,7 @@ class SIMPITIKI(datasets.GeneratorBasedBuilder):
|
|
201 |
name='challenge_seen_transformations_test',
|
202 |
# These kwargs will be passed to _generate_examples
|
203 |
gen_kwargs={
|
204 |
-
"filepath": downloaded_files['
|
205 |
"split": "challenge_seen_transformations_test",
|
206 |
},
|
207 |
),
|
@@ -210,7 +209,7 @@ class SIMPITIKI(datasets.GeneratorBasedBuilder):
|
|
210 |
name='challenge_unseen_transformations_test',
|
211 |
# These kwargs will be passed to _generate_examples
|
212 |
gen_kwargs={
|
213 |
-
"filepath": downloaded_files['
|
214 |
"split": "challenge_unseen_transformations_test",
|
215 |
},
|
216 |
),
|
@@ -263,10 +262,10 @@ class SIMPITIKI(datasets.GeneratorBasedBuilder):
|
|
263 |
data = json.load(f)
|
264 |
for id_, row in enumerate(data):
|
265 |
yield id_, {
|
266 |
-
"text":
|
267 |
-
"simplified_text":
|
268 |
-
"transformation_type":
|
269 |
-
"source_dataset":
|
270 |
"gem_id": f"gem-SIMPITIKI-{split}-{id_}",
|
271 |
}
|
272 |
|
|
|
61 |
"itwiki_test":"./v1/source_dataset_split/itwiki_test.json",
|
62 |
"tn_test":"./v1/source_dataset_split/tn_test.json"
|
63 |
}
|
|
|
|
|
64 |
},
|
65 |
"v2":{
|
66 |
"random": {
|
|
|
115 |
# This method specifies the datasets.DatasetInfo object which contains informations and typings for the dataset
|
116 |
features = datasets.Features(
|
117 |
{
|
118 |
+
"gem_id": datasets.Value("string"),
|
119 |
"text": datasets.Value("string"),
|
120 |
"simplified_text": datasets.Value("string"),
|
121 |
"transformation_type":datasets.Value("string"),
|
|
|
200 |
name='challenge_seen_transformations_test',
|
201 |
# These kwargs will be passed to _generate_examples
|
202 |
gen_kwargs={
|
203 |
+
"filepath": downloaded_files['transformations']['seen_transformations_test'],
|
204 |
"split": "challenge_seen_transformations_test",
|
205 |
},
|
206 |
),
|
|
|
209 |
name='challenge_unseen_transformations_test',
|
210 |
# These kwargs will be passed to _generate_examples
|
211 |
gen_kwargs={
|
212 |
+
"filepath": downloaded_files['transformations']['unseen_transformations_test'],
|
213 |
"split": "challenge_unseen_transformations_test",
|
214 |
},
|
215 |
),
|
|
|
262 |
data = json.load(f)
|
263 |
for id_, row in enumerate(data):
|
264 |
yield id_, {
|
265 |
+
"text": row["text"],
|
266 |
+
"simplified_text": row["simplified_text"],
|
267 |
+
"transformation_type":row["transformation_type"],
|
268 |
+
"source_dataset": row["source_dataset"],
|
269 |
"gem_id": f"gem-SIMPITIKI-{split}-{id_}",
|
270 |
}
|
271 |
|