Differentiable Rank
Types
Functions
SeqSpace.SoftRank.isotonic
— Methodisotonic(x)
Isotonically regress on data vector x
. Returns the monotonically increasing fit.
SeqSpace.SoftRank.partition
— Methodpartition(x; ϵ=1e-9)
Compute the sizes of each partition of constant value obtained by isotonic regression of data x
. ϵ
denotes the tolerance for what is considered equal in floating point terms.
SeqSpace.SoftRank.projection
— Methodprojection(x)
Regularize the ranking algorithm by euclidean projection onto the permutehedron.
SeqSpace.SoftRank.rank
— Methodrank(x)
Rank the items of vector x
in ascending order.
SeqSpace.SoftRank.softrank
— Methodsoftrank(x)
Regularization of the ranking algorithm. Scaled euclidean projection onto the permutehedron.
SeqSpace.SoftRank.∇isotonic
— Method∇isotonic(x)
Compute the gradient of isotonic regression with respect to the inputs about the solution point.