A bi-partite generative model framework for analyzing and simulating large scale multiple discrete-continuous travel behaviour data

•Data-driven interpretable machine learning for large-scale multiple discrete-continuous data.•Analytical methods for conditional probability, elasticity, latent variables, and fore-casting.•Incorporate variational Bayesian inference into travel behaviour model estimation.•Validation performed on a dataset consisting over 293,330 trip trajectories from Montréal.