Referencje:
pociąg
Użyj następującego polecenia, aby załadować ten zestaw danych do TFDS:
ds = tfds.load('huggingface:ubuntu_dialogs_corpus/train')
- Opis :
Ubuntu Dialogue Corpus, a dataset containing almost 1 million multi-turn dialogues, with a total of over 7 million utterances and 100 million words. This provides a unique resource for research into building dialogue managers based on neural language models that can make use of large amounts of unlabeled data. The dataset has both the multi-turn property of conversations in the Dialog State Tracking Challenge datasets, and the unstructured nature of interactions from microblog services such as Twitter.
- Licencja : Brak znanej licencji
- Wersja : 2.0.0
- Podziały :
Podział | Przykłady |
---|---|
'train' | 1000000 |
- Cechy :
{
"Context": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Utterance": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Label": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
test_programisty
Użyj następującego polecenia, aby załadować ten zestaw danych do TFDS:
ds = tfds.load('huggingface:ubuntu_dialogs_corpus/dev_test')
- Opis :
Ubuntu Dialogue Corpus, a dataset containing almost 1 million multi-turn dialogues, with a total of over 7 million utterances and 100 million words. This provides a unique resource for research into building dialogue managers based on neural language models that can make use of large amounts of unlabeled data. The dataset has both the multi-turn property of conversations in the Dialog State Tracking Challenge datasets, and the unstructured nature of interactions from microblog services such as Twitter.
- Licencja : Brak znanej licencji
- Wersja : 2.0.0
- Podziały :
Podział | Przykłady |
---|---|
'test' | 18920 |
'validation' | 19560 |
- Cechy :
{
"Context": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Ground Truth Utterance": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Distractor_0": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Distractor_1": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Distractor_2": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Distractor_3": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Distractor_4": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Distractor_5": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Distractor_6": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Distractor_7": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"Distractor_8": {
"dtype": "string",
"id": null,
"_type": "Value"
}
}