Hacktoberfest 2026: le issue che i maintainer hanno segnato per ottobre, aperte e adatte ai principianti. Sfoglia le issue Hacktoberfest

MCCD v1.3.6.3 vs PSFEx v1.4.6.3: shared inputs and outputs (masks, covariances)

Chiusa
  • Qualcosa non va?
#418 2 commenti 0 reazioni 0 assegnatari Vedi su GitHub

I maintainer di solito rispondono entro 1 giorno

Nessuno ha ancora preso questa issue.

Valutazione

Difficoltà
1/5
Tempo stimato
Meno di un'ora
Idoneità per principianti
52/100
Tipo di issue
Documentazione
Chiarezza
Abbastanza chiara
Stato di attività
Attiva
Stack tecnologico
python
Ambito
data

Direzione di ricerca

Start with the linked wiki results and the handoff README at /n17data/cdaley/unions/analyses/shear_2d/bmodes_2d/mccd_v1363_inputs/handoff/v1.3.6.3_blind_A/README.md; the issue provides data locations and caveats but does not specify a repository change. Determine what shared inputs and outputs need to be documented in the project, then confirm the relevant deliverable with the project maintainers before considering the work done.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Descrizione

Inputs and outputs from the Paper II B-mode pipeline runs on MCCD v1.3.6.3 vs PSFEx v1.4.6.3, so they can be reused (e.g. for E-mode work). Results: wiki. All paths are on candide and world-readable.

v1.3.6.3 handoff (for E-mode use): /n17data/cdaley/unions/analyses/shear_2d/bmodes_2d/mccd_v1363_inputs/handoff/v1.3.6.3/ — layout, grids and caveats are in its README.md.

  • Harmonic inference file cosmosis_SP_v1.3.6.3_leak_corr.fits, built with the same recipe as the v1.4.6.3 fiducial cosmosis_SP_v1.4.6.3_leak_corr_B.fits. That recipe rebuilds the v1.4.6.3 fiducial file byte-for-byte from its inputs.

    • COVMAT: iNKA Gaussian plus a OneCovariance NG term computed at the v1.3.6.3 area, mask and n(z). The combined covariance is pseudo_cl_cov_g_ng_iNKA_SP_v1.3.6.3_leak_corr.fits.
    • Build scripts: mccd_v1363_inputs/scripts/inference/.

  • Configuration + harmonic file cosmosis_SP_v1.3.6.3_leak_corr_xi_cl.fits, in the layout of the v1.4.6 combined file. It holds the reporting-binned ξ± (20 log bins, 1′–250′) with the cov_reporting_ng.txt covariance, plus the same Cℓ EE, covariance and n(z) as the harmonic file. ξ and Cℓ have separate covariances, so it is not a joint likelihood. It has no τ/ρ PSF-leakage HDUs. Build: mccd_v1363_inputs/combined/.

  • CosmoCov ξ± covariances. Each text matrix is ξ+ then ξ−, with θ increasing. _ng = Gaussian + non-Gaussian.

    • cov_reporting_g.txt, cov_reporting_ng.txt: 20 log bins, 1′–250′ (40×40).
    • cov_integration_g.txt: 1000 log bins, 0.5′–300′.

  • NaMaster Cℓ covariances (cl_cov_reporting.fits, cl_cov_fine.fits): EE/EB/BE/BB blocks. Gaussian only; COVAR_EE_EE is the iNKA Gaussian.

  • Footprint mask (nside 4096), mask Cℓ, and measured survey properties cov_th_SP_v1.3.6.3.yaml:

    A [deg²] n_e [arcmin⁻²] σ_e
    2504.98 4.625 0.3786

  • Planck18. n(z) = v1.3.6.3, unblinded (nz_SP_v1.3.6.3.txt in the handoff dir): the som_w-weighted histogram of Anna's v1.3.6.3 SOM calibration, in the same 200-bin format as the v1.4.6.3 files.

  • CosmoCov does not apply the footprint boundary correction to the shape-noise term (a few % at most inside 12′–83′).

lc outputs, blind B (B-modes, PTEs, covariances; branch mccd-v1363), under /n17data/cdaley/unions/analyses/shear_2d/bmodes_2d/pure_eb/results/:

  • mccd_v1363/: v1.3.6.3.
  • psfex_mccd_footprint/: v1.4.6.3 cut to the 2491 deg² shared with v1.3.6.3. Its catalog and mask are in mccd_v1363_inputs/control/.
  • psfex_no_p7/: v1.4.6.3 without P7. Its catalog and mask are in mccd_v1363_inputs/no_p7/.

Diagnostics:

  • mccd_v1363_inputs/footprint_diag/ANSWER.md: why v1.3.6.3 covers ~400 deg² less (pointing coverage).
  • mccd_v1363_inputs/no_p7/QUICKLOOK.md: P7 attribution.

Claude (Opus) on behalf of Cail

🤖 Generated with Claude Code

https://claude.ai/code/session_01GtjhbcRRBuyG8zrMbzcRY7

Lingua principale
Python
Stelle
2
Fork
5
Merge medio
14h 40m
PR unite (30g)
28

Preparare l'ambiente

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di CosmoStat/sp_validation

Tutte le issue di CosmoStat/sp_validation

Issue simili

Altre issue su Python

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.