|
import datasets |
|
|
|
|
|
_DATA = ["One", "Two", "Three", "Four", "Five", "Six", "Seven", "Eight", "Nine", "Ten"] |
|
|
|
|
|
class NewDataset(datasets.GeneratorBasedBuilder): |
|
|
|
|
|
VERSION = datasets.Version("1.0.0") |
|
|
|
def _info(self): |
|
return datasets.DatasetInfo( |
|
|
|
features=datasets.Features( |
|
{ |
|
"text": datasets.Value("string"), |
|
"meta": {"file": datasets.Value("string")}, |
|
} |
|
), |
|
|
|
|
|
|
|
) |
|
|
|
def _split_generators(self, dl_manager): |
|
return [ |
|
datasets.SplitGenerator( |
|
name=datasets.Split.TRAIN, |
|
gen_kwargs={ |
|
"data": _DATA, |
|
}, |
|
) |
|
] |
|
|
|
def _generate_examples(self, data): |
|
key = 0 |
|
for text in data: |
|
yield key, {"text": text, "meta": {"file": f"dir/file_{key + 1}.txt"}} |
|
key += 1 |
|
|
