ActivitySim Visualizer turns ActivitySim output into an interactive dashboard. Use it to examine one model run, compare multiple runs, or compare model results with survey data.
ActivitySim Visualizer can:
- prepare and summarize household, person, tour, and trip output;
- compare travel patterns, model choices, and validation measures for multiple runs;
- reuse valid cached results for faster startup; and
- launch a local dashboard or create a standalone HTML file.
From the repository root, use uv to create the environment and install the
locked dependencies:
uv sync --lockedIf Windows reports a hardlink problem, use:
uv sync --locked --link-mode=copyCopy config.yaml to local_config.yaml, then set each runs.dir value to an
ActivitySim output directory:
runs:
- dir: C:\models\base\output
label: Base
- dir: C:\models\build\output
label: BuildThe default file names are final_households, final_persons, final_tours,
final_trips, final_joint_tour_participants, and final_land_use. The
visualizer accepts CSV and Parquet input files.
For a small example configuration and instructions for nonstandard files or zones, see Getting Started and Configuring Your Data.
uv run activitysim-viz --config local_config.yamlOn the first run, the visualizer prepares the input, builds the required summary
tables, and starts a local server at
http://localhost:5006. Later runs reuse valid caches.
Press Ctrl+C to stop the server.
If data is missing or the first run fails, see Troubleshooting.
ActivitySim outputs
-> prepare canonical tables
-> summarize travel measures
-> display a live dashboard or export standalone HTML
The YAML configuration selects the input, workflow steps, output location, and dashboard mode. The start command stays the same for every workflow.
| Goal | Where to learn more |
|---|---|
| Use raw ActivitySim output directories | Configuring Your Data |
| Use already-prepared tables | Already-Prepared Tables |
| Use dashboard-ready summary tables | Dashboard-Ready Summary Tables |
| Run only the processor | Processor-Only Workflow |
| Create a standalone HTML dashboard | HTML Export |
| Understand caches and workflow steps | Running Workflows |
| Build summaries for configured subsets | Segmentation |
| Add district, county, or other zone groupings | Geography |
| Find an exact configuration field | Configuration Reference |
| Verify raw or prepared table requirements | Input Data Contract |
| Interpret cache manifests and rebuild decisions | Cache And Manifest Reference |
| Understand a summary table or field | Summary Catalog |
The wiki home is the main documentation index.
For a standard setup, read these chapters in order:
Other user references:
- Output Visualizer explains the dashboard.
- Dashboard User Guide lists the available analyses and explains how to interpret them.
- Posit Connect Cloud explains how to publish a standalone dashboard with the free public plan.
- Input Data Contract defines source and canonical table boundaries.
- Cache And Manifest Reference explains stored identities and diagnostics.
- HTML Export explains how to create an offline file.
- Summary Catalog documents every summary table.
- Segmentation explains subset summaries and their caches.
- Geography explains zone mappings and spatial outputs.
- Glossary defines project terminology.
- Troubleshooting covers common failures.
Start with Architecture and Developer Workflows, then use the guide for your task:
- extending prepared data;
- adding a summary function;
- adding dashboard pages, figures, or widgets;
- changing configuration, columns, or labels;
- skim enrichment; and
- testing.
Run focused tests during development. To run the full test suite, use:
uv run pytest --basetemp .pytest_tmpIf you change summary declarations or dashboard page definitions, regenerate the wiki catalogs from the code:
uv run python scripts/generate_wiki_catalogs.pyThe GNU General Public License v3.0 applies to this project. See
LICENSE.txt.