tomaarsen/

gooaq

This dataset is a collection of question-answer pairs, collected from Google. See GooAQ for additional information. This dataset can be used directly with Sentence Transformers to train embedding models. This dataset is equivalent to sentence-transformers/gooaq, but with 2500 evaluation samples and 2500 test samples randomly extracted from the training set.

	Dataset Subsets




	


	data subset

Columns: "question", "answer" Column types:… See the full description on the dataset page: https://huggingface.co/datasets/tomaarsen/gooaq.