qp_planck

QuickPol-based beam window function computation for Planck NPIPE.

This package wraps a lightly adapted version of the Planck NPIPE QuickPol pipeline to:

• build effective beam matrices for Planck detector / detset combinations;
• read RIMO (reduced instrument model) files for LFI and HFI;
• read or generate beam multipoles (Bₗₘ);
• compress the result into NPZ beam matrix files;
• convert those NPZ files into FITS Bℓ / Wℓ window functions.

---

Disclaimer

This code is adapted from the Planck NPIPE pipeline:

https://github.com/planck-npipe/toast-npipe/tree/master

The original implementation relied on TOAST and internal Planck tooling. Here, the code has been minimally modified to remove TOAST dependencies and to run as a standalone Python package.

---

Installation

You can install qp_planck locally for development:

git clone https://github.com/paganol/qp_planck.git
cd qp_planck
pip install -e .

This will install all Python modules in editable mode.
Make sure you have the following dependencies available:

• numpy
• scipy
• healpy
• astropy
• pyyaml
• mpi4py (optional, only if using MPI execution)

---

Package layout

After installation, the main package is:

qp_planck/
    qp_hmap2mat.py    # QuickPol driver: builds beam matrices and NPZ files (map → matrix)
    qp_mat2fits.py    # NPZ → FITS (Bℓ / Wℓ) window-function converter
    qp_pipeline.py    # High-level wrapper: runs hmap2mat + mat2fits; supports YAML configs
    utilities.py      # RIMO loading, detector lists, filename helpers, weights
    __init__.py       # Public API: load_RIMO, list_planck, qp_file,
                      #             hmap2mat, mat2fits, run_qp_pipeline
scripts/
    example_planck.yaml          # Example YAML configuration for running the full pipeline
    make_planck_detpairs.py      # Helper for generating Planck detset / detector pairs

---

Running modes

You can run the pipeline in three distinct ways.

1. Run using a YAML configuration (recommended)

You can drive the full pipeline (maps → matrices → FITS windows) with one command:

python -m qp_planck.qp_pipeline scripts/example_planck.yaml

The YAML file controls:

• list of detector pairs
• path to RIMO files
• blm file pattern
• output directory
• smax, lmax, release tag
• whether to run hmap2mat, mat2fits, or both

This is the cleanest way to reproduce full Planck-level outputs.

---

2. Run directly from Python

from qp_planck import run_qp_pipeline

run_qp_pipeline(
    detpairs=[("100GHz", "100GHz")],
    rimo_lfi="path/to/RIMO_LFI.fits",
    rimo_hfi="path/to/RIMO_HFI.fits",
    outdir="output/",
    run_hmap2mat=True,
    run_mat2fits=True
)

---

3. CLI: run low-level steps manually

Generate NPZ beam matrices:

python -m qp_planck.qp_hmap2mat

Convert NPZ → FITS:

python -m qp_planck.qp_mat2fits

These follow the original Planck QuickPol tools, with file naming via qp_file.

---

References

If you use this package, please cite:

QuickPol
E. Hivon, S. Mottet, N. Ponthieu,
QuickPol: Fast calculation of effective beam matrices for CMB polarization, A&A 598, A25 (2017)

Planck NPIPE
Planck Collaboration Int. LVII, NPIPE data processing pipeline, A&A 643, A42 (2020)