joecastiglione
commented
11 months ago SANDAG: SFCTA has done good visulation work using model results, can we build on top of what Joe has done? Addtionally, I think this is realted to the data model item 39. It would be great feature benefit all of us.
MTC: Interested in examples, technology options. (We have been using Tableau to quickly explore output but it's not free.)
Consolidating a number of Visualization features requests into this issue.
Visualization of Results
Related to the activitysim QC visualization feature, out-of-the-box components to (automatically) visualize model results would be quite useful to model users, developers, and consuming of our work in general. As the ActivitySim platform continues to mature and the user base grows, features like this will likely become increasingly important. The visualization software could be Python code or it could be a third party program such as one of the data visualization tools on the market or in the open source community. ActivitySim could include the ability to export data and reports in visualization-ready form. Related, a cloud-based version of activitysim would almost certainly require a reporting/visualization interface.
ActivitySim QC visualization
Several ad hoc reporting, validation, and testing scripts in Python and R have been developed to validate, verify, and debug ActivitySim model results. It would be good to build an ActivitySim QA/QC/reporting framework that is included with the software and can be used by users and developers. This would likely take the form of Python software with support for creating data summaries, maps, charts, etc. as well. With a framework in place, the model builder could implement several QA checks to ensure the model is behaving as expected. As the ActivitySim platform continues to mature and the user base grows, features like this will likely become increasingly important.
PopulationSim QC visualization
Redo the R validation script included with the example as a Python / Jupyter notebook. Include a new report model step as part of a run if desired. The benefit of this feature is to include built-in automated QA/QC procedures for improved usability. https://github.com/ActivitySim/populationsim/issues/92
"Met Council: the R validation scripts have been useful to the Met Council in comparing PopulationSIm to our previous population synthesizer. It might be interesting to explore if adding some additional visualizations might be useful, e.g. allowing users to set criteria for highlighting TAZs and markets segments that don't meet certain control total thresholds. Also interested in tools that might help users evaluate how one sythetic population scenario dffers from another to gain insight into model senstivity to demographic inputs.
SANDAG: this relies on a data model as described in item 39?
SEMCOG: We rewrote this R validation script in Jupyter notebook in order for us to compare the performance of SEMCOG's Synthpop and PopulationSim. "
SANDAG: SFCTA has done good visulation work using model results, can we build on top of what Joe has done? Addtionally, I think this is realted to the data model item 39. It would be great feature benefit all of us.
MTC: Interested in examples, technology options. (We have been using Tableau to quickly explore output but it's not free.)