Referencias:
largo
Utilice el siguiente comando para cargar este conjunto de datos en TFDS:
ds = tfds.load('huggingface:lccc/large')
- Descripción :
LCCC: Large-scale Cleaned Chinese Conversation corpus (LCCC) is a large corpus of Chinese conversations.
A rigorous data cleaning pipeline is designed to ensure the quality of the corpus.
This pipeline involves a set of rules and several classifier-based filters.
Noises such as offensive or sensitive words, special symbols, emojis,
grammatically incorrect sentences, and incoherent conversations are filtered.
- Licencia : MIT
- Versión : 1.0.0
- Divisiones :
| Separar | Ejemplos |
|---|---|
'train' | 12007759 |
- Características :
{
"dialog": [
{
"dtype": "string",
"id": null,
"_type": "Value"
}
]
}
base
Utilice el siguiente comando para cargar este conjunto de datos en TFDS:
ds = tfds.load('huggingface:lccc/base')
- Descripción :
LCCC: Large-scale Cleaned Chinese Conversation corpus (LCCC) is a large corpus of Chinese conversations.
A rigorous data cleaning pipeline is designed to ensure the quality of the corpus.
This pipeline involves a set of rules and several classifier-based filters.
Noises such as offensive or sensitive words, special symbols, emojis,
grammatically incorrect sentences, and incoherent conversations are filtered.
- Licencia : MIT
- Versión : 1.0.0
- Divisiones :
| Separar | Ejemplos |
|---|---|
'test' | 10000 |
'train' | 6820506 |
'validation' | 20000 |
- Características :
{
"dialog": [
{
"dtype": "string",
"id": null,
"_type": "Value"
}
]
}