-
Notifications
You must be signed in to change notification settings - Fork 100
Project Meeting 2016.06.15
bstabler edited this page Jan 15, 2018
·
8 revisions
- Logging and tracing are now implemented and we will review an example during the meeting
- For the trace HH, every data table - chooser, expressions/utilities, probabilities, choices, etc. - is written out to CSV file
- The order of trace output files is written to a master trace file
- Tests will be implemented by Friday
- Also fixed a bug so small sample sizes can now be run
- Add to issue some ideas for improvements
- Fix a couple small items that will be added to the issue as well
- Update all documentation for previous commits + logging and tracing
- Return to debugging/fixing the existing code for a full model run? (see below)
- We have delivered everything scoped + more for task 2 and task 3 and are just about done with task 9
- We have spent the entire budget for tasks 2, 3, and 9 as well
- We should return to debugging/fixing the existing code for a full model run but we don't have a task for this so can we re-allocate funds from other tasks?
- Skim handling is basically done, but we can't really confirm until we fix the inherited issues
- Can these issues be dealt with under the Nested Logit task for example? @bstabler: A bit, but we could implement a nested logit model without dealing with these inherited issues. It doesn't make a lot of sense to build upon a platform with these issues; let's correct the problems and then move forward
- MPO partners not surprised but need to discuss and get back to us
- Three options to consider: 1) add funding, 2) swap new task for existing tasks, 3) ignore and move forward with scoped tasks
- RSG will work on finishing tracing and updating documentation in the meantime
- Project schedule is fine
- MPO partners will also discuss their plan for next year as well
- Agreement with AMPO valid through June 2017
- Jeff will be half-time for the next two months while he vacations/works in France