Getting Started¶
This page is a quick on-ramp for new users: from installing Galacticus to running your first model and looking at its output.
1. Install Galacticus¶
Choose an installation route in Installation. The quickest is the pre-compiled binary; if you intend to modify the code, build from source instead.
2. Set up the environment¶
Galacticus needs two environment variables — GALACTICUS_EXEC_PATH (the
Galacticus directory) and GALACTICUS_DATA_PATH (the run-time datasets) — as
described in Running Galacticus. The run-time datasets are downloaded separately; see
the installation instructions for your platform.
3. Run your first model¶
Galacticus is run by passing it a parameter file:
Galacticus.exe parameters/quickTest.xml
quickTest.xml is a small, fast model included with the source. A successful
run writes its results to an HDF5 file named galacticus.hdf5 in the working
directory. Running Galacticus walks through this run step by step and shows how to
explore the output.
What is a parameter file?¶
Everything about a Galacticus model — the physics included, the cosmology, and what to output — is controlled by an XML parameter file. A minimal example looks like:
<parameters>
<outputFileName value="myModel.hdf5"/>
<!-- ... further parameters ... -->
</parameters>
Each element sets one parameter. Many parameters select a class implementation for a piece of physics — for example, which halo mass function or cooling model to use. The Advanced Usage chapter describes parameter files in depth: conditional parameters, math expressions, sub-parameters, and how the output is structured. The full list of available parameters for every class is documented in the physics reference.
4. Analyze the output¶
Galacticus output is standard HDF5 and can be read from any language. The
companion Dendros package provides
ready-made analysis and plotting tools, including on-the-fly analyses, MCMC chain
diagnostics, and posterior corner plots (pip install dendros).
Next steps¶
Running Galacticus — a detailed walk-through of running a model and reading its output.
Advanced Usage — parameter files, output structure, and advanced usage.
Troubleshooting — what to do if something goes wrong.
A collection of tutorials is available on the wiki.