A pytorch extension library to perform triple walks on knowledge graphs
Project description
A simple algorithm to learn embeddings of entities in knowledge graph.
What is it
TripleWalk is an algorithm to learn vector embeddings of entities in a knowledge graph by performing random walks on triples.
Installation
Please not this package is only available for python 3.8+ and Pytorch >= 1.9.0
Install from PyPI
pip install triple-walk
Install from Github
pip install git+https://github.com/udel-cbcb/triple_walk.git#egg=triple_walk
Triple Walk
Author : Sachin Gavali
Requirements
1. Pytorch >= 1.9.0
2. NVIDIA-GPU (Cuda Toolkit >= 11.4
3. AMD-GPU (ROCM == 4.0.1)
4. Python == 3.8
Examples
- SkipGram Triple Walk model : SkipGramTriple
- CBOW Triple Walk model : CBOWTriple
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
triple_walk-0.0.4.tar.gz
(19.4 kB
view hashes)