Welcome to NXTfusion’s documentation!

NXTfusion is a pytorch-based library for non-linear data fusion over Entity-Relation graph.

• What is Data Fusion?
  • WHAT IS NXTfusion
  • What is this an Entity-Relation graph?
  • NXTfusion approach generalizes existing data fusion methods
  • Examples from the scientific world
  • What is this repository for?
• Installation guide
  • How do I set it up?
  • Install from git repo
  • Install from pip
  • Install from conda
• Quickstart
  • Example1: single (nonlinear) matrix factorization
• Advanced examples
  • Example2: tensor factorization
  • Example3: Two relations among 3 entities
  • Example4: Using side information
• How to build a NN model to be used in NXTfusion
  • NN model for single matrix factorization
  • A NN for tensor factorization
  • A NN for inference over arbitrary ER graphs
  • Further reading
• List of modules and functions in nxtfusion
  • NXTfusion.DataMatrix module
  • NXTfusion.NXFeaturesConstruction module
  • NXTfusion.NXLosses module
  • NXTfusion.NXTfusion module
  • NXTfusion.NXmodels module
  • NXTfusion.NXmultiRelSide module
• How to cite
• About us
• Disclaimer

Indices and tables

• Index

• Module Index

• Search Page