From Voice To Shell: Text To Luminaires Dataset

Dataset with 1,272 distinct samples of commands for an emercengy luminaires environment along with their expression as natural language, called 'prompt'. English (EN) version is comprised by three splits: train (795 samples), validation (371 samples), and test (106 samples). Spanish (ES) version holds only a test split (106 samples).本数据集包含1272条适用于应急照明环境的独立命令样本，以及各命令对应的自然语言表述，该表述被称为“提示（prompt）”。其英文版本划分为训练集（795个样本）、验证集（371个样本）与测试集（106个样本）三个子集；西班牙语版本仅包含测试子集，共计106个样本。