https://github.com/mnick/holographic-embeddings

Code for experiments in the AAAI 2016 paper "Holographic Embeddings of Knowledge Graphs"
https://github.com/mnick/holographic-embeddings

Last synced: 2 days ago
JSON representation

Code for experiments in the AAAI 2016 paper "Holographic Embeddings of Knowledge Graphs"

• Host: GitHub
• URL: https://github.com/mnick/holographic-embeddings
• Owner: mnick
• Created: 2015-11-21T21:16:09.000Z (almost 9 years ago)
• Default Branch: master
• Last Pushed: 2018-08-13T22:53:05.000Z (over 6 years ago)
• Last Synced: 2024-08-02T13:22:30.979Z (3 months ago)
• Language: Python
• Homepage:
• Size: 1.37 MB
• Stars: 173
• Watchers: 22
• Forks: 51
• Open Issues: 8
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Holographic Embeddings of Knowledge Graphs

This repository holds the code for experiments in the paper 

```

Holographic Embeddings of Knowledge Graphs

Maximilian Nickel, Lorenzo Rosasco, Tomaso Poggio, AAAI 2016.

```

## Install 

To run the experiments, first install [scikit-kge](https://github.com/mnick/scikit-kge),

An open-source python library to compute various knowledge graph embeddings including

- Holographic Embeddings (HolE)

- RESCAL

- TransE

- TransR

- ER-MLP

After `scikit-kge` is installed, simply clone this repository via 

```

git clone [email protected]:mnick/holographic-embeddings.git

```

and run the experiments as detailed in the next section

## Experiments 

The repository holds scripts of the form 

```

run__.sh

```

which runs the experiments for `dataset` with the best parameters for `model`.

The full code for the experiments can be found in the `kg` and `countries` subfolders. The python scripts in these subfolders should be easy to use for grid search.