Each time you do a project in Python, it is highly recommended to create a virtual python environment to keep the base python environment of your system 'clean', and easily allow you to switch version if needed. This time is no different. To prevent issues regarding the installation of python packages, we highly recommend creating a virtual environment of version 3.10.13. For instructions on how to do this, have a look at popular python environment mangagers such as conda, pyenv, and PyFlow.
Once the virtual environment has been created and activated, the necesary package can be installed using either $conda install --file requirements.txt when using conda, $pip install -r requirements.txt for pyenv, or $pyflow init for PyFlow.
To use the newly created virtual python environment in Jupyter Notebook;
- move to the right directory,
- open a terminal,
- activate the virtual environment (i.e.
$conda activate [YOUR_VENV]or$pyenv activate [YOUR_VENV]if you are using conda or pyenv. PyFlow should do this automatically), - and start a local jupyter server (i.e.
$jupyter notebook). Instructions on how to access the server should be printed in the terminal.
Feel free to reach out to the TAs if you have any questions!
part 1: RDF
Go through the notebook rdf-tutorial
part 2: RDFS
Go through the notebook rdfs-tutorial
part 3: OWL
Try out the notebook OWL-tutorial
part 4: SPARQL
Notebook SPARQL-tutorial
part 5: RDFstar
Notebook RDFstar-tutorial
part 6: Machine Learning over KGs
Notebook ML4KG-tutorial
part 7: Entity Linking and Relation Extraction
The ./data/ingredients.rdf and ./data/recipes.rdf are based on the github repo: https://github.com/foodkg/foodkg.github.io.git
The ./data/musicoset_metadata/ files are downloaded from: https://marianaossilva.github.io/DSW2019/#tables