common_gen

Tài liệu tham khảo:

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:common_gen')
  • Sự miêu tả :
CommonGen is a constrained text generation task, associated with a benchmark dataset, 
to explicitly test machines for the ability of generative commonsense reasoning. Given 
a set of common concepts; the task is to generate a coherent sentence describing an 
everyday scenario using these concepts.

CommonGen is challenging because it inherently requires 1) relational reasoning using 
background commonsense knowledge, and 2) compositional generalization ability to work 
on unseen concept combinations. Our dataset, constructed through a combination of 
crowd-sourcing from AMT and existing caption corpora, consists of 30k concept-sets and 
50k sentences in total.
  • Giấy phép : Không có giấy phép được biết đến
  • Phiên bản : 2020.5.30
  • Chia tách :
Tách ra Ví dụ
'test' 1497
'train' 67389
'validation' 4018
  • Đặc trưng :
{
    "concept_set_idx": {
        "dtype": "int32",
        "id": null,
        "_type": "Value"
    },
    "concepts": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "target": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}