Data and Code for Basian Spatial Interaction Model (BSIM).
Python 3.6.4
TensorFlow 2.2.0
The data that support the findings of this study was generated by combining open source and commercial proprietary data. The commercial proprietary data were obtained from the industrial partner to support research purposes and are subject to strict non-disclosure agreements. However, researchers interested in replicating our results on the commercial problems can directly request data from the relevant organisations using the references made in the paper. The data used for synthetic experiments and data created from the open sources for real-world applications are provided under `Data'.
The synthetic experiments are conducted under both Variational Inference (VI) and Markov chain Monte Carlo (MCMC). VI approach can be executed by running Synthetic.py in Experiments folder and MCMC method can be executed with the MCMC_Stan.R code in MCMC folder.
The real experiments are only implemented with VI method since the scalability. This can be executed by running Real.py in the Experiments folder. Please note that only the data acquired from the open source is given in the Data folder and other fields are kept blank.