Which loss functions are suitable for sentence ordering?

[ghost]

Given the features are tokenized sentences, and the targets y_pred are normalized rankings. A typical model accepts tokenized sentences as inputs and outputs their order/ranks.

`x`: tokenize(['sentence 1', 'sentence 2', 'sentence 3', 'sentence 4'])
`y_true`: [0., 0.33333333, 0.66666667, 1.]

Which of the loss function implementations is suitable for this kind of data?

[PrzemekPobrotyn]

If I understood your setup correctly I think all implemented loss functions should work. Depending if you choose pointwise/ pairwise or listwise loss you predict the normalised ranking directly / compare two sentences and learn to order them correctly / learn to order the entire list at once, respectively