lm1b

הפניות:

טקסט_פשוט

השתמש בפקודה הבאה כדי לטעון מערך נתונים זה ב-TFDS:

ds = tfds.load('huggingface:lm1b/plain_text')
  • תיאור :
A benchmark corpus to be used for measuring progress in statistical language modeling. This has almost one billion words in the training data.
  • רישיון : אין רישיון ידוע
  • גרסה : 1.0.0
  • פיצולים :
לְפַצֵל דוגמאות
'test' 306688
'train' 30301028
  • מאפיינים :
{
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}