Datasets:
Tasks:
Text Classification
Sub-tasks:
sentiment-classification
Languages:
English
Size:
1K - 10K
ArXiv:
License:
Update etpc.py
Browse files
etpc.py
CHANGED
|
@@ -87,6 +87,12 @@ class ETPC(datasets.GeneratorBasedBuilder):
|
|
| 87 |
"sentence2_segment_location": datasets.Sequence(
|
| 88 |
datasets.Value("int32")
|
| 89 |
),
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 90 |
"sentence1_segment_text": datasets.Sequence(
|
| 91 |
datasets.Value("string")
|
| 92 |
),
|
|
@@ -116,7 +122,6 @@ class ETPC(datasets.GeneratorBasedBuilder):
|
|
| 116 |
]
|
| 117 |
|
| 118 |
def _generate_examples(self, file_paths):
|
| 119 |
-
|
| 120 |
file_paths = list(file_paths)
|
| 121 |
text_pairs_path = file_paths[0]
|
| 122 |
paraphrase_types_path = file_paths[1]
|
|
@@ -168,6 +173,9 @@ class ETPC(datasets.GeneratorBasedBuilder):
|
|
| 168 |
len(sentence2_tokenized)
|
| 169 |
)
|
| 170 |
|
|
|
|
|
|
|
|
|
|
| 171 |
for (
|
| 172 |
sentence1_segment_locations,
|
| 173 |
sentence2_segment_locations,
|
|
@@ -180,6 +188,7 @@ class ETPC(datasets.GeneratorBasedBuilder):
|
|
| 180 |
segment_locations_1 = [
|
| 181 |
int(i) for i in sentence1_segment_locations.split(",")
|
| 182 |
]
|
|
|
|
| 183 |
sentence1_segment_location_full[segment_locations_1] = [
|
| 184 |
paraphrase_type_id
|
| 185 |
] * len(segment_locations_1)
|
|
@@ -187,6 +196,7 @@ class ETPC(datasets.GeneratorBasedBuilder):
|
|
| 187 |
segment_locations_2 = [
|
| 188 |
int(i) for i in sentence2_segment_locations.split(",")
|
| 189 |
]
|
|
|
|
| 190 |
sentence2_segment_location_full[segment_locations_2] = [
|
| 191 |
paraphrase_type_id
|
| 192 |
] * len(segment_locations_2)
|
|
@@ -204,6 +214,8 @@ class ETPC(datasets.GeneratorBasedBuilder):
|
|
| 204 |
"paraphrase_type_ids": paraphrase_type_ids,
|
| 205 |
"sentence1_segment_location": sentence1_segment_location_full,
|
| 206 |
"sentence2_segment_location": sentence2_segment_location_full,
|
|
|
|
|
|
|
| 207 |
"sentence1_segment_text": sentence1_segment_text,
|
| 208 |
"sentence2_segment_text": sentence2_segment_text,
|
| 209 |
}
|
|
|
|
| 87 |
"sentence2_segment_location": datasets.Sequence(
|
| 88 |
datasets.Value("int32")
|
| 89 |
),
|
| 90 |
+
"sentence1_segment_location_indices": datasets.Sequence(
|
| 91 |
+
datasets.Sequence(datasets.Value("int32"))
|
| 92 |
+
),
|
| 93 |
+
"sentence2_segment_location_indices": datasets.Sequence(
|
| 94 |
+
datasets.Sequence(datasets.Value("int32"))
|
| 95 |
+
),
|
| 96 |
"sentence1_segment_text": datasets.Sequence(
|
| 97 |
datasets.Value("string")
|
| 98 |
),
|
|
|
|
| 122 |
]
|
| 123 |
|
| 124 |
def _generate_examples(self, file_paths):
|
|
|
|
| 125 |
file_paths = list(file_paths)
|
| 126 |
text_pairs_path = file_paths[0]
|
| 127 |
paraphrase_types_path = file_paths[1]
|
|
|
|
| 173 |
len(sentence2_tokenized)
|
| 174 |
)
|
| 175 |
|
| 176 |
+
sentence1_segment_indices = []
|
| 177 |
+
sentence2_segment_indices = []
|
| 178 |
+
|
| 179 |
for (
|
| 180 |
sentence1_segment_locations,
|
| 181 |
sentence2_segment_locations,
|
|
|
|
| 188 |
segment_locations_1 = [
|
| 189 |
int(i) for i in sentence1_segment_locations.split(",")
|
| 190 |
]
|
| 191 |
+
sentence1_segment_indices.append(segment_locations_1)
|
| 192 |
sentence1_segment_location_full[segment_locations_1] = [
|
| 193 |
paraphrase_type_id
|
| 194 |
] * len(segment_locations_1)
|
|
|
|
| 196 |
segment_locations_2 = [
|
| 197 |
int(i) for i in sentence2_segment_locations.split(",")
|
| 198 |
]
|
| 199 |
+
sentence2_segment_indices.append(segment_locations_2)
|
| 200 |
sentence2_segment_location_full[segment_locations_2] = [
|
| 201 |
paraphrase_type_id
|
| 202 |
] * len(segment_locations_2)
|
|
|
|
| 214 |
"paraphrase_type_ids": paraphrase_type_ids,
|
| 215 |
"sentence1_segment_location": sentence1_segment_location_full,
|
| 216 |
"sentence2_segment_location": sentence2_segment_location_full,
|
| 217 |
+
"sentence1_segment_location_indices": sentence1_segment_indices,
|
| 218 |
+
"sentence2_segment_location_indices": sentence2_segment_indices,
|
| 219 |
"sentence1_segment_text": sentence1_segment_text,
|
| 220 |
"sentence2_segment_text": sentence2_segment_text,
|
| 221 |
}
|