This code is written to fulfill the requirements of my bachelor thesis
The Objective is to fine-tune models for the ORKG completion.
The approach is simple; I prepared ORKG data and format it to a QA with context dataset and use it to train models already pretrained using the Squad v2 dataset
first prepare the dataset -> for this run the ./data_preparation/main.py second run the ./training_evaluation/main.py to train and evaluate the models
Experience with python might be needed to understand and run the script
All the dependencies are described in the ./requirements.txt you can use 'pip install -r requirements.txt' to install the dependencies The script was written using Python version 3.10 (other version might also be compatible)
Also, you need to run "python -m spacy download en_core_web_sm" to download the spacy pipeline we use in the data preparation part of the script
Follow the ./data_preparation/README.md file and the ./training_evaluation/README.md file
This service is developed by: Moussab Hrou, moussab.hrou@stud.uni-hannover.de
The code in ./data_preparation/fetch_abstracts.py is from link