jpwahle commited on
Commit
40ea8f9
1 Parent(s): 27753d4

Update etpc.py

Browse files
Files changed (1) hide show
  1. etpc.py +13 -1
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
  }