English | Español

Get the corpus running: install dependencies, run scenarios, run the analysis pipeline, and open the dashboard.

See Installation for step-by-step setup (clone ONE, compile, Python venv, verify).

• Java (e.g. OpenJDK 11+) — for The ONE simulator
• The ONE — built from source (compile with ./compile.sh in the ONE root)
• Python 3 with: numpy, pandas, scipy, matplotlib, streamlit
• Repository — either the scenario corpus only, or the full ONE repo with scenarios/ inside it (or linked)

Typical layout:

• ONE root (e.g. the-one/ or the-one-scenario-corpus/) — contains one.sh, compile.sh, default_settings.txt, reports/ (where ONE writes report files)
• scenarios/ — contains corpus_v1/, analysis/, README, ROADMAP, and this wiki content

If the scenario corpus is a separate repo, run the ONE from its repo and point scripts to the path where you cloned the scenarios (e.g. --corpus /path/to/scenarios/corpus_v1).

Run a single scenario (from the ONE root):

./one.sh -b 1 scenarios/corpus_v1/01_urban/U1_CBD_Commuting_HelsinkiMedium.settings

Run all 60 scenarios (from the ONE root, or from the repo that contains scenarios/):

python3 scenarios/analysis/run_all_scenarios.py --corpus corpus_v1

• Reports (e.g. *_MessageStatsReport.txt) are written to the directory set in each .settings (often reports/ at the ONE root).
• To only list scenarios without running: add --dry-run.

From the repository root (parent of scenarios/):

# Extract features → analysis/data/features.csv
python3 scenarios/analysis/run_analysis.py --corpus corpus_v1 --phase features

# Normalise (z-score) → analysis/data/features_normalized.csv
python3 scenarios/analysis/run_analysis.py --phase normalize

# Correlation matrices and reports → analysis/data/*.csv, analysis/reports/*.txt
python3 scenarios/analysis/run_analysis.py --phase correlation

# Figures → analysis/figures/*.png, *.pdf
python3 scenarios/analysis/run_analysis.py --phase figures

# Build output_metrics.csv from ONE reports (if reports/ exists)
python3 scenarios/analysis/run_analysis.py --phase output_metrics

# Output-based correlation (needs output_metrics.csv)
python3 scenarios/analysis/run_analysis.py --phase outputs

Run everything (features through output_metrics) in one go:

python3 scenarios/analysis/run_analysis.py --corpus corpus_v1 --phase all

Run outputs separately after you have analysis/data/output_metrics.csv (from ONE runs + output_metrics phase).

From the repository root:

streamlit run scenarios/analysis/dashboard.py

• Opens a browser with: summary, results by phase, per-scenario view, scenario comparison.
• Requires streamlit and pandas (and the same Python env as the pipeline).

• Java/ONE not found — Run from the directory that contains one.sh; ensure ONE is compiled.
• Python module not found — Use a venv and install dependencies: pip install numpy pandas scipy matplotlib streamlit.
• Corpus path — Use --corpus corpus_v1 if you run from the repo root and corpus_v1 is at scenarios/corpus_v1. Otherwise use the full path to the corpus directory.
• output_metrics phase — Needs ONE report files (e.g. *_MessageStatsReport.txt) in the reports directory; run scenarios first or set --reports-dir if reports are elsewhere.

• Installation — Step-by-step setup
• Reproducibility — Regenerating analysis from scratch
• Methodology — How features and correlation work
• Results overview — Main results and figures
• Corpus overview — Scenario families and design