Closed sloth2012 closed 5 years ago
Yes, I have consider this. Check this function in the code. We use a set to store the triplets. Therefore, {(1,2,3),(4,5,6)} is the same with {(4,5,6), (1,2,3)}.
@xiangrongzeng Sorry, i didn't notice that. thank you!
Have you considered that relation facts order will affect the loss count? for example: target triplets may be like this: [(1,2,3),(4,5,6)], which should be equal to [(4,5,6), (1,2,3)]. with the way counting loss that paper mentioned, it may be quit different.