id_panl_bppt

सन्दर्भ:

• कोड
• गले लगाना

id_panl_bppt

इस डेटासेट को TFDS में लोड करने के लिए निम्नलिखित कमांड का उपयोग करें:

ds = tfds.load('huggingface:id_panl_bppt/id_panl_bppt')

• विवरण :

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).

• लाइसेंस : कोई ज्ञात लाइसेंस नहीं
• संस्करण : 1.0.0
• विभाजन :

विभाजित करना	उदाहरण
'train'	24021

• विशेषताएँ :

{
    "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"
    }
}