Skip to content
RonBibbPublic

About

Unpolarized Gowdy spike-sector test finding a converged localized mismatch for four-function VTD data while leaving the ordinary region architecture-limited.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Latest commit

 

History

1 Commit

Folders and files

Repository files navigation

BB-CBE-3: Unpolarized Gowdy Spike-Sector Test

This project tests the four-function velocity-dominated reconstruction in unpolarized Gowdy T³ data containing localized spike candidates. It is the first adverse stage in the Line-E collapse–bang equivalence program.

Result

Aggregate: intermediate/CBE-D with a converged adverse localized spike result.

  • Localized spike candidates show a resolution-consistent mismatch of approximately D ≈ 0.82 under the four-function VTD reconstruction.
  • The ordinary region remains formally unclassified because the original mask is contaminated by the broader spike footprint and the backward reconstruction is ill-conditioned.
  • Transient spikes were absent in the production data.
  • Later exact-family controls showed that the mismatch diagnoses information missing from the four-function data set, not necessarily failure of equivalence under an enriched spike state.

Contents

  • run_cbe3.py — unpolarized evolution, detector, reconstruction, and scoring.
  • CBE3.json — complete machine-readable classification.
  • cbe3_run_log*.txt — preserved execution logs.

Reproduction

Requires Python 3 with NumPy and SciPy:

python3 run_cbe3.py

Scope

The detector classified localized candidates; this repository alone does not certify exact Lim-family membership or almost-everywhere equivalence.

About

Unpolarized Gowdy spike-sector test finding a converged localized mismatch for four-function VTD data while leaving the ordinary region architecture-limited.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages