We started the hackathon as a team of four - two techies and two sociologists. After a full day of setup, our sociologist couldn't even take a look at the spatial data to bring in their knowledge about social influences of land prices, which is what we needed to build a prediction model around. With frustration about the painful setup, they gave up.
To make data exploration easier for our fallen friends, we made it our mission to build a data literacy dashboard so that anyone can explore spatial data without much setup.
Our dashboard will show you every insight you want to know about the land prices in Bremen. From basic questions like 'How is the density of young people in each neighborhood?' to more advanced questions like 'How much does the distance from the center contribute to the land price in the neighborhood Altbremen?', you can answer them all.
We built it using the provided data together with Streamlit, XGBoost and Shapley values.
Losing two of our teammates and all the setup that the spatial data needed (a lot of data, new programs and libraries).
Getting everything to work in such a short time and learning all these new geospatial tools, as well as streamlit!
Thanks for organizing the hackathon - Nico and Morris
Install all dependencies in conda env
To run the streamlit web app, run
streamlit run main.py
To run the model evaluation, go to folder challenge_evaluation
cd challenge_evaluation
python model.py
- Morris Kurz
- Nicholas Wolf