lm1b

참고자료:

일반 텍스트

TFDS에 이 데이터세트를 로드하려면 다음 명령어를 사용하세요.

ds = tfds.load('huggingface:lm1b/plain_text')
  • 설명 :
A benchmark corpus to be used for measuring progress in statistical language modeling. This has almost one billion words in the training data.
  • 라이센스 : 알려진 라이센스 없음
  • 버전 : 1.0.0
  • 분할 :
나뉘다
'test' 306688
'train' 30301028
  • 특징 :
{
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}