Extend input pipeline to tomography#191
Draft
cailmdaley wants to merge 8 commits into
Draft
Conversation
script: resolves through the workflow symlink into sp_validation, where docs/talks/ doesn't exist. shell: runs from pure_eb root where it does. Also remove container: None so plots run inside the container. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Results-table generator: \hline -> \hline\hline at top, add \noalign{\vskip 3pt}
after each horizontal rule. Matches A&A house style (double-rule top, single-rule
bottom, with breathing room).
- Remove bold_threshold=0.05 from standalone PTE macros (\cosebisfiducialPte,
\ebfiducialPte, and all per-version \{prefix\}Xip/Xim/Combined/Cosebis variants).
Bolding now only applies to table cells (which use _pte_row_cells with an
explicit bold_threshold), not to single-value PTEs embedded in prose.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The UNIONS-3500 series was renumbered earlier in 2026 (Hervas-Peters et al. catalog = Paper I, this work = Paper II, Goh et al. config = Paper III, Guerrini et al. harmonic = Paper IV, Hervas-Peters et al. sims = Paper V). The config-space spec docs still referenced the old numbering. Switch to topic-based names ("B-modes paper", "config-space paper", "harmonic paper") so future renumbering doesn't cascade.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
plotting_utils: - Add ell_bin_mask, ell_bin_index, get_powspace_bin_edges helpers. A bin is now included in a scale-cut PTE only if its full multipole range lies within [ell_min, ell_max] — replacing the naive nearest-bin selection that off-by-one'd. - PTE colormap rebuilt with BoundaryNorm + ListedColormap for exact threshold breaks at 0.05/0.95 (4096 gradient levels between). New PTE_COLORBAR_TICKS, PTE_COLORBAR_TICKLABELS, format_pte_colorbar() helper. Consumers switched to the helpers: - cl_data_vector.py, cl_version_comparison.py: use ell_bin_mask in _compute_pte_with_cuts; render shading via actual bin edges; legend repositioned. - harmonic_space_pte_matrices.py: use ell_bin_index for fiducial PTE indexing (this is the off-by-one fix Douglas Scott caught in Fig 7's black-square position). Fontsizes bumped 7→9 across labels; "drop 1 tick" colorbar override (cosmetic — avoids 0.95/1 collision in the appendix composite). claims.smk: - COSEBIS fiducial angular range now uses geomspace bin edges (250**(9/20), 250**(16/20) ≈ 12, 83 arcmin) to match the config_space_pte_matrices grid exactly. Drops the silent inconsistency between paper-quoted scale cuts and the cuts actually used in cross-path COSEBIS PTE computations. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
cosebis_data_vector + harmonic_config_cosebis_comparison: - Figure labels/titles: "COSEBIS" → "COSEBI" (singular form preferred throughout the paper). harmonic_config_cosebis_comparison: - Cap reliable mode count at 6 for both 32-bin and 96-bin configurations (was 6 / 8). At 96 bins, modes 1-6 recover to <0.3% on GLASS mocks; higher modes are progressively less reliable due to W_n(ell) oscillation resolution. 6 also matches the mode range used for cosmological inference throughout the paper. - Column titles in the combined comparison figure now read "All scales" and "Fiducial scale cuts" instead of raw "θ = X-Y'" — easier on the reader. - Drop the sharey=row constraint so each row scales independently. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two new rules + scripts for the GLASS-mock COSEBIS bias test: - mock_cosebis_scatter (per-mock): reads fine-binned xi±, applies scale cuts, computes COSEBIS E_n/B_n. Byte-order conversion for numba compatibility handled inside the script. - mock_cosebis_bias_test (gather): collects per-mock COSEBIS from the scatter jobs, propagates CosmoCov xi± covariance into COSEBIS space, tests mean B_n = 0 at σ/√N precision. Two-panel figure: B_n/σ_analytic for each mock (faded) + mean with σ/√N error bars on top; σ_empirical / σ_analytic per mode on bottom. GLASS_MOCK_IDS bumped from 5 to 100 (full mock set now available). glass_mock_xi_fine resources adjusted: 48→24 threads, 30→20 GB, 120→180 min runtime — accommodates the larger mock set without over-requesting. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Snakefile: include rules/presentation.smk (file was committed earlier but the top-level include wasn't wired in). plot_presentation_blind_nz.py: plot the three blinded n(z) curves used in the Moriond slides. Single-panel, three colored lines, light grid. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stub — extend the input/analysis pipeline to tomographic bins. Follows the restructuring (#188).
— Claude on behalf of Cail
🤖 Generated with Claude Code