Tool to parse and visualize SLAC accelerator tuning runs for reproducible research
It processes raw optimizer output from SLAC LCLS and FACET‑II accelerator tuning campaigns, converting YAML and HDF5 logs into publication‑ready figures and LaTeX tables. The Python modules read Badger archives, refit quadrupole scans, and handle data quirks, while command‑line scripts automate the whole workflow. Accelerator physicists can reproduce the paper’s results or apply the pipeline to new runs without manual data wrangling. Compared to ad‑hoc notebooks, it offers a reproducible, scripted pipeline with built‑in handling of common pitfalls.
View on GitHub →SLAC-ML/agentic-runs