pygrank
A Python library bringing together fast recommendation algorithms for large graphs based on link analysis.
Developed by
License
Apache License 2.0 (Apache-2.0)
Main Characteristic
- Compatibility with networkx, numpy, tensorflow, and pytorch.
- Datacentric interfaces that do not require transformations to identifiers.
- Large graph support with sparse representations and fast algorithms.
- Seamless pipelines, from graph preprocessing up to benchmarking and evaluation.
- Modular combination of components.
Technical Categories
AI services
Last updated
12.04.2022 - 14:18
Detailed Description
This is an open source Python package to define, run and evaluate node ranking algorithms. It provides object-oriented and extensively unit-tested algorithmic components, such as graph filters, postprocessors, measures, benchmarks and online tuning. These can be combined to define interoperable complex algorithms. Computations can be delegated to numpy or GPU backends and fit in back-propagation pipelines.