GLaMoR

This Repository contains all the code required to run the GLaMoR-Pipeline.

Project Structure

├──GLaMoR-DataPipeline
│  ├──Data_Retrieval/ 
│  ├──Preprocessing/
│  ├──Initial_Publish/
│  ├──OAPT/
│  ├──OWL_Ontology_Modification/
│  ├──Prefix Removal/
│  ├──Translation/
│  ├──Tokenization/
│  ├──Embed/
│  ├──Analysis/
├──GLaMoR-Model_Training
│  ├──Data_Format_Translation/
│  ├──GLaMoR-HermiT/
│  ├──GLaMoR-ModernBert/
│  ├──GraphLanguageModels/
│  ├──Llama/
│  ├──Logistic_Regression/
│  ├──OWL2Vec-Star/
│  ├──Prodigy/
│  ├──WideMLP
├──.gitignore/
├──.gitmodules
├──README.md
├──LICENSE

Requirements

The System is split into two components, the Data Pipeline and the Model Training. The requirements for the two components are listed below

Data Pipeline

If you want to use the code as provided, it is enough to have docker-compose installed on the system and run

> docker-compose build 
> docker-compose up -d 

in the root directory. Remember to update the mountings as well as environment variables according to your setup.

Model Training

To train the models, a requirements.txt file is available for most of them. If it's missing, make sure you have the following installed:

• scikit-learn
• pandas
• numpy
• torch

Citation

A citation will be provided upon publication. In the meantime, please contact justin.muecke@uni-ulm.de for more information.