Are there any methods to visualize simulation runs from a batch run after the fact? I would like to view the spatial layout of agents from different runs at the last tick. I think freezedrying would work, but it is a very large complex system. Once the system gets large enough, trying to load the default xml file results in a Java Heap error. Ideally, I only need class type and spatial location saved. I'm unsure how to customize the freezedrying process in this manner, any guidance?
This is a great question. What we usually do is to output the minimal information needed to construct our visualizations (rather than freezedrying, which is more for checkpointing purposes) and use scripts written in R or Python to do this. For your case you could consider outputting CSV files listing the class type and spatial locations.