Repository navigation
137: per-triangle DEM height histogram (--dem-heights) - #431
Merged
Merged
Conversation
…area; DEM pixel statistics argued out, their walk held (@architect) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…out its plane, behind --dem-heights; plan area dropped (@architect) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…hreshold, the reprojected path read on the source's own pixels (@architect) Design review of 6f0e20d returned CHANGES REQUESTED: the 50-node threshold left 72 % of Bygdin's triangles without a histogram. Measured: the overlap walk costs 1.2 times the error summary's node walk at 10 threads on Bygdin, no extra peak memory; the resampled grid narrows histograms by 4-8 % in spread against the source's pixels, so the source is read. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…8fe80c, float32 tolerances) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…erance wide, the residual at the pixel centre (@architect) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
… residual named rho) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…lished vertical accuracy, each triangle carrying only the classes its own nodes reach (@architect) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…GLO-30's 2 m LE90 as 1.216 m, for ANADEM too; DTM10's absolute 1 m kept as an upper bound) (@architect) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…ise, Velhas agreement, size +507, row-sum bound, convergence range (@architect) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…angles) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…eld nodes' span, classes as wide as the DEM's relative accuracy (@tester) Binding (tests 4-8, 19, the margin, edges, refusals, GIL, stub), dem_heights and HeightTable (tests 3, 22, chunks, reprojected), the reader (test 15), the command (tests 1, 2, 10, 11, 13, 14, 16-18, 20, 21, --triangles), the off digests at 98fe80c keyed by platform (test 12), the C++ property (test 9); the new field joins the catalogue tests, the module the layering table, the binding the threads list. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…m=None, last, on purpose for 137 (@tester) Increment 137 (3.4) adds the field to the source model; plugins see one more keyword with a None default. The PR body says so. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…its plane, sparse by its held nodes' span, in .vtk, .ply and .gpkg (@developer) C++: statistics/height_shares.hpp on overlap.hpp's clip, now a piece visitor shared with class_areas (its bytes unchanged); the binding dem_height_shares. Python: dem_heights (classes, chunks, the reprojected path on the source's pixels), HeightTable, the writers and the reader, relative_vertical_accuracy_m on sources (1.216 m for glo30 and anadem-v1), and --dem-heights with its refusals and record entries. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
… 4 are classes +-2, not the end classes (@tester) The redone kill record on a1d6c1d found the mutant that drops the division in class_of's quotient (height_shares.hpp:65) alive: the raise and lower steps repair every class except where the early clamp fires, which needs a residual past n + 1 m but inside (n + 1/2) w, so w > 1. No scene had one. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…s n = 8, though the quotient's floor says 7 (@tester) The redone kill record on a1d6c1d found the mutant that raises n while (n + 1/2) w < tolerance, not <= (dem_heights.py:46), alive: the floor of tolerance / w + 1/2 is already right on every listed case, so the raise step was never exercised. At w = 1.216 the quotient rounds low on the edge, as in test 4's residual case. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…corded Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…bytes; on costs +0.07 s and +10 MB on the fixture tile, +0.03 s on a 4-piece cut, nothing on Velhas (@Perf) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
… - 1/2) w, and the kill record is docs/increments/137-probes/kill-record.md (@tester) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…-party section (ruff I001) (@tester) Co-Authored-By: Claude Opus 5.5 <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
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.
Increment 137 (docs/increments/137-dem-stats.md): each triangle's DEM height histogram behind
--dem-heights, off by default.577 counted lines. Red fa7cc0e, plus 237447f for plugin_api SIGNATURES, which that test asks to be updated on its own. Green a1d6c1d. Kill record: 37 mutants over 14 targets plus three extra, all killed or proven equivalent (docs/increments/137-probes/kill-record.md). @Perf section 8: with the flag off, time and bytes equal master; with it on, +0.065 s and +10 MB on the test tile.
Review verdicts:
🤖 Generated with Claude Code