def _info(self):
# TODO(xquad): Specifies the datasets.DatasetInfo object
return datasets.DatasetInfo(
# This is the description that will appear on the datasets page.
description=_DESCRIPTION,
# datasets.features.FeatureConnectors
features=datasets.Features({
"id":
datasets.Value("string"),
"context":
datasets.Value("string"),
"question":
datasets.Value("string"),
"answers":
datasets.features.Sequence({
"text":
datasets.Value("string"),
"answer_start":
datasets.Value("int32"),
}),
# These are the features of your dataset like images, labels ...
}),
# If there's a common (input, target) tuple from the features,
# specify them here. They'll be used if as_supervised=True in
# builder.as_dataset.
supervised_keys=None,
# Homepage of the dataset for documentation
homepage="https://github.com/deepmind/xquad",
citation=_CITATION,
task_templates=[
QuestionAnsweringExtractive(question_column="question",
context_column="context",
answers_column="answers")
],
)
def test_column_mapping(self):
task = QuestionAnsweringExtractive(
context_column="input_context", question_column="input_question", answers_column="input_answers"
)
self.assertDictEqual(
{"input_context": "context", "input_question": "question", "input_answers": "answers"}, task.column_mapping
)
def _info(self):
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=datasets.Features({
"id":
datasets.Value("string"),
"title":
datasets.Value("string"),
"context":
datasets.Value("string"),
"question":
datasets.Value("string"),
"answers":
datasets.features.Sequence({
"text":
datasets.Value("string"),
"answer_start":
datasets.Value("int32"),
}),
}),
# No default supervised_keys (as we have to pass both question
# and context as input).
supervised_keys=None,
homepage="https://rajpurkar.github.io/SQuAD-explorer/",
citation=_CITATION,
task_templates=[
QuestionAnsweringExtractive(question_column="question",
context_column="context",
answers_column="answers")
],
)
def test_map_on_task_template(self):
info = DatasetInfo(task_templates=QuestionAnsweringExtractive())
dataset = Dataset.from_dict({k: [v] for k, v in SAMPLE_QUESTION_ANSWERING_EXTRACTIVE.items()}, info=info)
assert isinstance(dataset.info.task_templates, list)
assert len(dataset.info.task_templates) == 1
def keep_task(x):
return x
def dont_keep_task(x):
out = deepcopy(SAMPLE_QUESTION_ANSWERING_EXTRACTIVE)
out["answers"]["foobar"] = 0
return out
mapped_dataset = dataset.map(keep_task)
assert mapped_dataset.info.task_templates == dataset.info.task_templates
# reload from cache
mapped_dataset = dataset.map(keep_task)
assert mapped_dataset.info.task_templates == dataset.info.task_templates
mapped_dataset = dataset.map(dont_keep_task)
assert mapped_dataset.info.task_templates == []
# reload from cache
mapped_dataset = dataset.map(dont_keep_task)
assert mapped_dataset.info.task_templates == []
def _info(self):
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=datasets.Features(
{
"id": datasets.Value("int32"),
"title": datasets.Value("string"),
"context": datasets.Value("string"),
"question": datasets.Value("string"),
"answers": datasets.features.Sequence(
{
"text": datasets.Value("string"),
"answer_start": datasets.Value("int32"),
}
),
}
),
supervised_keys=None,
homepage="",
citation=_CITATION,
task_templates=[
QuestionAnsweringExtractive(
question_column="question", context_column="context", answers_column="answers"
)
],
)
def _info(self):
return ds.DatasetInfo(
description="",
citation="",
homepage="",
license="",
features=ds.Features({
"id":
ds.Value("string"),
"title":
ds.Value("string"),
"context":
ds.Value("string"),
"question":
ds.Value("string"),
"answers":
ds.features.Sequence({
"text": ds.Value("string"),
"answer_start": ds.Value("int32")
}),
}),
task_templates=[
QuestionAnsweringExtractive(question_column="question",
context_column="context",
answers_column="answers")
],
)
def _info(self):
features = datasets.Features({
"document_id":
datasets.Value("int32"),
"context":
datasets.Value("string"),
"question":
datasets.Value("string"),
"is_impossible":
datasets.Value("bool"),
"id":
datasets.Value("int32"),
"answers":
datasets.features.Sequence({
"text": datasets.Value("string"),
"answer_start": datasets.Value("int32"),
}),
})
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=features,
supervised_keys=None,
homepage=_HOMEPAGE,
license=_LICENSE,
citation=_CITATION,
task_templates=[
QuestionAnsweringExtractive(question_column="question",
context_column="context",
answers_column="answers")
],
)
def _info(self):
features = datasets.Features({
"squad_v2_id":
datasets.Value("string"),
"original question":
datasets.Value("string"),
"disfluent question":
datasets.Value("string"),
"title":
datasets.Value("string"),
"context":
datasets.Value("string"),
"answers":
datasets.features.Sequence({
"text": datasets.Value("string"),
"answer_start": datasets.Value("int32"),
}),
})
return datasets.DatasetInfo(
# This is the description that will appear on the datasets page.
description=_DESCRIPTION,
# This defines the different columns of the dataset and their types
features=
features, # Here we define them above because they are different between the two configurations
# If there's a common (input, target) tuple from the features,
# specify them here. They'll be used if as_supervised=True in
# builder.as_dataset.
supervised_keys=None,
# Homepage of the dataset for documentation
homepage=_HOMEPAGE,
# License for the dataset if available
license=_LICENSE,
# Citation for the dataset
citation=_CITATION,
task_templates=[
QuestionAnsweringExtractive(
question_column="disfluent question",
context_column="context",
answers_column="answers")
],
)
def test_from_dict(self):
input_schema = Features({
"question": Value("string"),
"context": Value("string")
})
label_schema = Features({
"answers":
Sequence({
"text": Value("string"),
"answer_start": Value("int32"),
})
})
template_dict = {
"context_column": "input_input_context",
"question_column": "input_question",
"answers_column": "input_answers",
}
task = QuestionAnsweringExtractive.from_dict(template_dict)
self.assertEqual("question-answering-extractive", task.task)
self.assertEqual(input_schema, task.input_schema)
self.assertEqual(label_schema, task.label_schema)
def _info(self):
# TODO(tydiqa): Specifies the datasets.DatasetInfo object
if self.config.name == "primary_task":
return datasets.DatasetInfo(
# This is the description that will appear on the datasets page.
description=_DESCRIPTION,
# datasets.features.FeatureConnectors
features=datasets.Features({
"passage_answer_candidates":
datasets.features.Sequence({
"plaintext_start_byte":
datasets.Value("int32"),
"plaintext_end_byte":
datasets.Value("int32"),
}),
"question_text":
datasets.Value("string"),
"document_title":
datasets.Value("string"),
"language":
datasets.Value("string"),
"annotations":
datasets.features.Sequence({
# 'annotation_id': datasets.Value('variant'),
"passage_answer_candidate_index":
datasets.Value("int32"),
"minimal_answers_start_byte":
datasets.Value("int32"),
"minimal_answers_end_byte":
datasets.Value("int32"),
"yes_no_answer":
datasets.Value("string"),
}),
"document_plaintext":
datasets.Value("string"),
# 'example_id': datasets.Value('variant'),
"document_url":
datasets.Value("string")
# These are the features of your dataset like images, labels ...
}),
# If there's a common (input, target) tuple from the features,
# specify them here. They'll be used if as_supervised=True in
# builder.as_dataset.
supervised_keys=None,
# Homepage of the dataset for documentation
homepage="https://github.com/google-research-datasets/tydiqa",
citation=_CITATION,
)
elif self.config.name == "secondary_task":
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=datasets.Features({
"id":
datasets.Value("string"),
"title":
datasets.Value("string"),
"context":
datasets.Value("string"),
"question":
datasets.Value("string"),
"answers":
datasets.features.Sequence({
"text":
datasets.Value("string"),
"answer_start":
datasets.Value("int32"),
}),
}),
# No default supervised_keys (as we have to pass both question
# and context as input).
supervised_keys=None,
homepage="https://github.com/google-research-datasets/tydiqa",
citation=_CITATION,
task_templates=[
QuestionAnsweringExtractive(question_column="question",
context_column="context",
answers_column="answers")
],
)
