id_panl_bppt

참고자료:

id_panl_bppt

TFDS에 이 데이터세트를 로드하려면 다음 명령어를 사용하세요.

ds = tfds.load('huggingface:id_panl_bppt/id_panl_bppt')
  • 설명 :
Parallel Text Corpora for Multi-Domain Translation System created by BPPT (Indonesian Agency for the Assessment and
Application of Technology) for PAN Localization Project (A Regional Initiative to Develop Local Language Computing
Capacity in Asia). The dataset contains around 24K sentences divided in 4 difference topics (Economic, international,
Science and Technology and Sport).
  • 라이센스 : 알려진 라이센스 없음
  • 버전 : 1.0.0
  • 분할 :
나뉘다
'train' 24021
  • 특징 :
{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "translation": {
        "languages": [
            "en",
            "id"
        ],
        "id": null,
        "_type": "Translation"
    },
    "topic": {
        "num_classes": 4,
        "names": [
            "Economy",
            "International",
            "Science",
            "Sport"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}