I had to create my conda environment to run the code. I used pipreqs to scan the project that outputs the following requirements.txt file:
matplotlib==3.9.0 numpy==2.0.0 pandas==2.2.2 plotly==5.22.0
scipy==1.13.1 seaborn==0.13.2
As mentioned in the paper, I commented pystan and installed it afterward with version 2.17.1.0. Cython was needed for the installation as well.
I am not very familiar with the statistical procedures/tools used in the paper. I am familiar with Python programming language.
I used Miniconda to create my conda environment.
I used Miniconda and vscode to run the Jupyter notebooks. In the environment, I installed Python 3.9 and the dependencies listed by pipreqs with specific version for pystan and Cython
The main challenge was creating the environment to run the code.
The Jupiter notebooks format was an easy setup to reproduce the results.
I will suggest providing a requeriments.txt file with the specific versions used in the project. Ideally providing a docker image.
Providing a requirements.txt file with the necessary dependencies.
I like the simplicity and the fact that the folder structure in the project is well-defined and explained.