References:
gutenberg
Use the following command to load this dataset in TFDS:
ds = tfds.load('huggingface:gutenberg_time/gutenberg')
- Description:
A clean data resource containing all explicit time references in a dataset of 52,183 novels whose full text is available via Project Gutenberg.
- License: [More Information needed]
- Version: 0.0.0
- Splits:
Split | Examples |
---|---|
'train' |
120694 |
- Features:
{
"guten_id": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"hour_reference": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"time_phrase": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"is_ambiguous": {
"dtype": "bool_",
"id": null,
"_type": "Value"
},
"time_pos_start": {
"dtype": "int64",
"id": null,
"_type": "Value"
},
"time_pos_end": {
"dtype": "int64",
"id": null,
"_type": "Value"
},
"tok_context": {
"dtype": "string",
"id": null,
"_type": "Value"
}
}