Skip to content

dice-embeddings 1.0.5
Compare
Choose a tag to compare
@Demirrr Demirrr released this 20 Jan 12:38
· 915 commits to main since this release
v1.0.5
9460079

Features

  1. An AbstractCallback class is implemented. Few callback classes are created to print info related to training, to save the model paramaters and apply polyak parameter ensemble model PrintCallback, KGESaveCallback, PPE.
  2. Pandas, Modin and Polars can be used as a backend. Reading, preprocessing, saving and loading can be done in a parallel fashion.
  3. AccumulateEpochLossCallback
  4. Gradient Accumulation is implemented.

Applications

  1. A function for predicting conjunctive queries over knowledge graph is implemented.
  2. A function to detect missing triples is implemented.
Assets 2
Loading