Reference arrays (not tracked in Git)
This folder should contain the reference arrays used by validation:
- bkgd.npy, signal.npy
- bkgd_scores.npy, signal_scores.npy
- data_raw.npy, signal_raw.npy (or data_A_raw.npy, signal_WH_raw.npy)
- boundaries.npy, significances.npy
- file_list.txt, root_summary.txt
Why they are missing on GitHub:
- The repository's .gitignore excludes *.npy to avoid committing large binary files.
- These files are hundreds of MB and unsuitable for source control.
How to populate locally: You can copy from the shared path on NERSC:
/global/cfs/projectdirs/atlas/dwkim/llm_test_dev_cxyang/llm_for_analysis/solution/arrays
Or use the helper script at scripts/fetch_solution_arrays.sh from the repo root:
bash scripts/fetch_solution_arrays.sh
This will copy the required files into solution/arrays/.