All notable changes to PathSim are documented here.
The format follows Keep a Changelog. PathSim uses Semantic Versioning.
- Monte Carlo simulation engine with vectorised NumPy implementation
- Three built-in scenarios: startup, career-change, investment
- Distribution support: normal, beta, uniform
- Sensitivity analysis via Spearman rank correlation (no scipy dependency)
- CLI with
richformatting, progress spinner, and outcome bars --chartflag for matplotlib score histograms--explainflag for optional Ollama LLM explanation--seedflag for reproducible results--listflag to enumerate available scenarios- Natural-language decision matching (e.g.
"start a startup"→ startup scenario) - Programmatic API (
pathsim.engine.simulate) - Full pytest test suite
- GitHub Pages documentation site
- MIT license