data split

[fsh2017]

‘’‘ We split the dataset into training, validation, and test sets so that each set includes all recipe types, resulting in 2,923, 838, and 1,241 data for the training, validation, and test sets, respectively. ‘’‘

Thanks for your work. I noticed that the segmentation of your dataset doesn't conform to the classical paradigm. May I ask what the reason is for this division?