lm1b

อ้างอิง:

ข้อความธรรมดา

ใช้คำสั่งต่อไปนี้เพื่อโหลดชุดข้อมูลนี้ใน TFDS:

ds = tfds.load('huggingface:lm1b/plain_text')
  • คำอธิบาย :
A benchmark corpus to be used for measuring progress in statistical language modeling. This has almost one billion words in the training data.
  • ใบอนุญาต : ไม่มีใบอนุญาตที่รู้จัก
  • เวอร์ชัน : 1.0.0
  • แยก :
แยก ตัวอย่าง
'test' 306688
'train' 30301028
  • คุณสมบัติ :
{
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}