Project Meeting 2022.11.17

Agenda

• Code Review Updates
• Memory Profiling Updates

Meeting Notes

Code Review Updates

• School escorting and flexible number of tour and trip IDS
• Working on school escorting comments from Joe
• RSG address flexible number of tour and trip IDs from Joe
• Estimation and Random Seeds
• RSG has estimation fix that they are working on
• RSG submitted codes to Joe on random seeds
• Disaggregate Accessibilities
• Sijia provided comments that RSG will
• Sharrow
• RSG comments are in, done from their perspective
• WSP working on it

Memory Profiling

• Follow-up on action items
• Data type table – found a place in the settings yaml where users can specify the type of data. Tested with 25% ARC but not much memory savings (26GB memory reduction), less than saving that originally expected
• Just by updating input data type, doesn’t help that much
• When data is used in combination, model defaults to the higher precision, so that may be what’s going on
• ACTION ITEM: data type usage table, wanted to add in data rows for each of the files to get a sense of data size when you look at number of rows and precision
• Are there even ways to fix this efficiently? Joe remembered something that Jeff said about how difficult this might be, ACTION ITEM to check with Jeff on this.
• Still to upload tableau workbook to tableau public
• Next steps
• Mandatory tour scheduling is biggest memory user and look into ways to reduce; options include:
• Reduce alternative set size that are very unlikely
• Look into what RSG did for SEMCOG for this component
• MWCOG example is crashing, something with the time window
• MWCOG to ask their staff to look into it
• They have a separate development repo that works
• They will run that to confirm it’s error free and then send to Sijia
• MWCOG could share their latest code, even if it’s not checked in
• Sijia to contact Feng Xie
• Discussion about chunking
• This was something implemented to solve a problem, was never an ideal solution
• Chunking addresses some initial design issues and some asked if we need to go back to the original design to see if there is a better implementation that wouldn’t require chunking
• Joe is wanting to look for lower hanging fruit first before considering a big refactoring process, for example

Other

• KPMG did a security scan on the ActivitySim code (for a Sydney project) and identified some issues, some of which are easy to fix and others require more work. report is shared here:DetailedReport_ActivitySim_1_Nov_2022.pdf