参考:
enwik8
使用以下命令在 TFDS 中加载此数据集:
ds = tfds.load('huggingface:enwik8/enwik8')
- 说明:
The dataset is based on the Hutter Prize (http://prize.hutter1.net) and contains the first 10^8 byte of Wikipedia
- 许可:无已知许可
- 版本:1.0.0
- 拆分:
拆分 | 样本 |
---|---|
'train' |
1128024 |
- 特征:
{
"text": {
"dtype": "string",
"id": null,
"_type": "Value"
}
}
enwik8-raw
使用以下命令在 TFDS 中加载此数据集:
ds = tfds.load('huggingface:enwik8/enwik8-raw')
- 说明:
The dataset is based on the Hutter Prize (http://prize.hutter1.net) and contains the first 10^8 byte of Wikipedia
- 许可:无已知许可
- 版本:1.0.0
- 拆分:
拆分 | 样本 |
---|---|
'train' |
1 |
- 特征:
{
"text": {
"dtype": "string",
"id": null,
"_type": "Value"
}
}