=============== Getting started =============== WS4 has two ways in, and most people need only one of them: * **The command line.** One command per step of a study, from compiling a dataset to writing a report. No Python, nothing to import. Start with :doc:`quickstart`. * **The Python API.** The same steps as library calls, for automation, custom diagnostics, sensitivity sweeps, and integration with other tools. Start with :doc:`python-api`. Both paths read and write the same files, and both can be mixed: a command-line run can be resumed in Python, and a Python study can shell out to the commands it does not want to reimplement. .. list-table:: :header-rows: 1 :widths: 26 74 * - Page - What it covers * - :doc:`installation` - Requirements, the optional extras, and a three-command install check. * - :doc:`quickstart` - The command-line walkthrough: compile, simulate, target, search, watch. * - :doc:`python-api` - The same walkthrough as Python, plus the in-memory evaluator and a hand-driven search. * - :doc:`first-model` - A complete small study, with the numbers checked by hand. * - :doc:`architecture` - How the parts fit together, and the pitfalls that follow from the design. .. toctree:: :maxdepth: 2 :hidden: installation quickstart python-api first-model architecture