docs(paper): update validation results after audit of the benchmarks - #165
Merged
Conversation
Audited the satellite packages' validation routines against the original TerraME Lua scripts. Two defects were found in the comparators, none in the models themselves, and the reported figures improve. BR-MANGUE: the metrics loop compared the final state against every golden CSV, and the golden files are indexed on state rather than executions (step_01.csv is the initial state). With both fixed, land use and soil match TerraME exactly at every checkpoint (MAE 0, max error 0) and elevation improves from 90.7% to 97.3% within 1 mm. Adds the flood component coverage caveat: the reference scenario triggers no land-use transition, and the golden files confirm TerraME does the same. Lab6: replaces kappa with the Pontius & Millones quantity/allocation decomposition, consistent with the Pontius reference already cited, and adds the discriminance caveat -- a trivial static ranking reproduces the same output, so the scenario validates coefficient transcription rather than the allocation algorithm. Lab1: replaces 'expected divergence between independent implementations' with the mechanism, since the LuccME script declares maxDifference=1643, a 7.6% convergence band within which the reference itself stops short of its declared demand. Notes the perturbation tests that confirm the benchmark constrains the implementation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Audited the satellite packages' validation routines against the original TerraME Lua scripts. Two defects were found in the comparators, none in the models themselves, and the reported figures improve.
BR-MANGUE: the metrics loop compared the final state against every golden CSV, and the golden files are indexed on state rather than executions (step_01.csv is the initial state). With both fixed, land use and soil match TerraME exactly at every checkpoint (MAE 0, max error 0) and elevation improves from 90.7% to 97.3% within 1 mm. Adds the flood component coverage caveat: the reference scenario triggers no land-use transition, and the golden files confirm TerraME does the same.
Lab6: replaces kappa with the Pontius & Millones quantity/allocation decomposition, consistent with the Pontius reference already cited, and adds the discriminance caveat -- a trivial static ranking reproduces the same output, so the scenario validates coefficient transcription rather than the allocation algorithm.
Lab1: replaces 'expected divergence between independent implementations' with the mechanism, since the LuccME script declares maxDifference=1643, a 7.6% convergence band within which the reference itself stops short of its declared demand. Notes the perturbation tests that confirm the benchmark constrains the implementation