Have a look at the Jupyter Notebook (.ipynb)
If you want to fully explore the model in the associated paper, you need to open and run the Jupyter Notebook in this repository. Have a look at the pdf and html files to have an idea of the expected output.
-
Clone the present repository in a folder of your choice
git clone https://github.com/grassoste/SP-secretion-efficiency-Jupyter-notebook.git
-
Create a suitable environemnt (Python > 3.6) and install relevant packages. If you prefer you can use your local installation of Python. In such case install the same packages but use 'pip' (i.e. pip install)
conda create -n SP-notebook python=3.6 conda activate SP-notebook cd ‘directory-of-notebook’ conda install jupyter conda install pandas conda install matplotlib conda install joblib conda install -c conda-forge shap jupyter notebook
-
Navigate to the Jupyter notebook and run it. Feel free to play around. We suggest to refer to the docs of SHAP for a better understanding: https://shap.readthedocs.io/en/latest/
Feel free to contact us if you run into issues: s [dot] grasso [at] umcg [dot] nl