JAS-mine maintains and develops the JAS simulation platform, a discrete-event tool-kit for agent-based and dynamic microsimulation modelling. This repository contains the core libraries. See www.jas-mine.net for more details.
Output from a multirun simulation is currently inconsistent. There seems to be confusion about how results ought to be saved. Line 191 of ExperimentManager, for example, is coded as dbFile.mkdir() - designed to create a new output directory for the database of a simulation. Inclusion of this line suggests use of a new output directory for each pass of a multirun. Yet a separate directory for each pass of a multirun is not created, so that the cited code fails (quietly - it could be made to work by recoding as dbFile.mkdirs()). Output from each multirun is currently appended to the csv files of the original run. If this is the preferred behaviour, then related code should be cleaned up.
Output from a multirun simulation is currently inconsistent. There seems to be confusion about how results ought to be saved. Line 191 of ExperimentManager, for example, is coded as dbFile.mkdir() - designed to create a new output directory for the database of a simulation. Inclusion of this line suggests use of a new output directory for each pass of a multirun. Yet a separate directory for each pass of a multirun is not created, so that the cited code fails (quietly - it could be made to work by recoding as dbFile.mkdirs()). Output from each multirun is currently appended to the csv files of the original run. If this is the preferred behaviour, then related code should be cleaned up.