Datasets:
Sebastian Gehrmann
commited on
Commit
•
43577d3
1
Parent(s):
9b7245a
fix naming
Browse files
RotoWire_English-German.py
CHANGED
@@ -145,6 +145,8 @@ class RotowireEnglishGerman(datasets.GeneratorBasedBuilder):
|
|
145 |
"summary_en": datasets.Sequence(datasets.Value("string")),
|
146 |
"sentence_end_index_en": datasets.Sequence(datasets.Value("int32")),
|
147 |
"summary_de": datasets.Sequence(datasets.Value("string")),
|
|
|
|
|
148 |
"sentence_end_index_de": datasets.Sequence(datasets.Value("int32")),
|
149 |
"linearized_input": datasets.Value("string")
|
150 |
}
|
@@ -465,7 +467,7 @@ class RotowireEnglishGerman(datasets.GeneratorBasedBuilder):
|
|
465 |
linearized_input = " ".join(output)
|
466 |
linearized_input = linearized_input.replace(" ", " ")
|
467 |
return linearized_input
|
468 |
-
|
469 |
def _generate_examples(
|
470 |
self, filepath, split # method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
471 |
):
|
@@ -477,4 +479,6 @@ class RotowireEnglishGerman(datasets.GeneratorBasedBuilder):
|
|
477 |
all_data = json.load(f)
|
478 |
for id_, data in enumerate(all_data):
|
479 |
data['linearized_input'] = self.linearize_input(data)
|
|
|
|
|
480 |
yield id_, data
|
|
|
145 |
"summary_en": datasets.Sequence(datasets.Value("string")),
|
146 |
"sentence_end_index_en": datasets.Sequence(datasets.Value("int32")),
|
147 |
"summary_de": datasets.Sequence(datasets.Value("string")),
|
148 |
+
"target": datasets.Value("string"),
|
149 |
+
"references": [datasets.Value("string")],
|
150 |
"sentence_end_index_de": datasets.Sequence(datasets.Value("int32")),
|
151 |
"linearized_input": datasets.Value("string")
|
152 |
}
|
|
|
467 |
linearized_input = " ".join(output)
|
468 |
linearized_input = linearized_input.replace(" ", " ")
|
469 |
return linearized_input
|
470 |
+
|
471 |
def _generate_examples(
|
472 |
self, filepath, split # method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
473 |
):
|
|
|
479 |
all_data = json.load(f)
|
480 |
for id_, data in enumerate(all_data):
|
481 |
data['linearized_input'] = self.linearize_input(data)
|
482 |
+
data['target'] = data['summary_de'][0]
|
483 |
+
data['references'] = data['summary_de']
|
484 |
yield id_, data
|