Quickstart logistic growth
This is a reproducible LawSynth example. It generates observations from the
model stated in config.toml, performs finite-difference
polynomial-library discovery where supported, and simulates the declared
world. All generated artifacts are written below output/ and are ignored by
version control.
python generate.py
python discover.py
python simulate.py
python -m pytest test_example.py
The data are synthetic and deterministic; they are suitable for exercising the pipeline, not for claiming causal identification from observational data.
The included standard-library baseline is intentionally transparent. If the
native lawsynth extension is installed, discovery invokes it as well and
records its actual outcome in output/discovery.json.