Crate tranz 

tranz

Point-embedding knowledge graph completion models.

Entities are points in vector space. Relations are transformations (translation, rotation, diagonal scaling). Lower distance between transform(head, relation) and tail indicates a more likely triple.

Models

• TransE: head + relation ~ tail (Bordes et al., 2013)
• RotatE: head * relation ~ tail in complex space (Sun et al., 2019)
• ComplEx: Hermitian dot product in complex space (Trouillon et al., 2016)
• DistMult: diagonal bilinear in real space (Yang et al., 2015)

Feature flags

• rand (default): enables random initialization via Model::new().
• candle: enables GPU training via the train module.
• cuda: implies candle, enables CUDA acceleration.

Modules

dataset

Dataset loading for KGE benchmark triple files.

eval

Link prediction evaluation with filtered ranking.

io

Embedding import/export.

train

Training loop for KGE models via candle.

Structs

ComplEx

ComplEx: complex bilinear model.

DistMult

DistMult: diagonal bilinear model in real space.

RotatE

RotatE: rotation in complex space.

TransE

TransE: translational distance model.

Enums

Error

Errors from tranz operations.

Traits

Scorer

Trait for scoring knowledge graph triples.