metadata
task_categories:
- conditional-text-generation
AutoTrain Dataset for project: dippatel_summarizer
Dataset Description
This dataset has been automatically processed by AutoTrain for project dippatel_summarizer.
Languages
The BCP-47 code for the dataset's language is unk.
Dataset Structure
Data Instances
A sample from this dataset looks as follows:
[
{
"feat_id": "13864393",
"text": "Peter: So have you gone to see the wedding?\nHolly: of course, it was so exciting\nRuby: I really don't understand what's so exciting about it\nAngela: me neither\nHolly: because it's the first person of colour in any Western royal family\nRuby: is she?\nPeter: it's not true\nHolly: no?\nPeter: there is a princess in Liechtenstein\nPeter: I think a few years ago a prince of Liechtenstein married a woman from Africa\nPeter: and it was the first case of this kind among European ruling dynasties\nHolly: what? I've never heard of it\nPeter: wait, I'll google it\nRuby: interesting\nPeter: here: <file_other>\nPeter: Princess Angela von Liechtenstein, born Angela Gisela Brown\nPeter: sorry, she's from Panama, but anyway of African descent\nRuby: right! but who cares about Liechtenstein?!\nPeter: lol, I just noticed that it's not true, what you wrote\nRuby: I'm excited anyway, she's the first in the UK for sure",
"target": "Holly went to see the royal wedding. Prince of Liechtenstein married a Panamanian woman of African descent."
},
{
"feat_id": "13716378",
"text": "Max: I'm so sorry Lucas. I don't know what got into me.\nLucas: .......\nLucas: I don't know either.\nMason: that was really fucked up Max\nMax: I know. I'm so sorry :(.\nLucas: I don't know, man.\nMason: what were you thinking??\nMax: I wasn't.\nMason: yea\nMax: Can we please meet and talk this through? Please.\nLucas: Ok. I'll think about it and let you know.\nMax: Thanks...",
"target": "Max is sorry about his behaviour so wants to meet up with Lucas and Mason. Lucas will let him know. "
}
]
Dataset Fields
The dataset has the following fields (also called "features"):
{
"feat_id": "Value(dtype='string', id=None)",
"text": "Value(dtype='string', id=None)",
"target": "Value(dtype='string', id=None)"
}
Dataset Splits
This dataset is split into a train and validation split. The split sizes are as follow:
| Split name | Num samples |
|---|---|
| train | 2400 |
| valid | 600 |