Research-Stack/5-Applications/tools-scripts/carrier/carrier_constants_ne.py

46 lines
2.9 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

# ==============================================================================
# COPYRIGHT NO ONE EVERYWHERE LLC (WYOMING HOLDING COMPANY)
# PROJECT: SOVEREIGN STACK
# This artifact is entirely proprietary and cryptographically proven.
# Open-Source usage requires explicit permission from Brandon Scott Schneider.
# ==============================================================================
# soliton_constants_ne.py — Non-Euclidean geometry constants
# Generated by: tools/calibrate_geometry.py
# DAG tick 767, geometry-rip branch
#
# Method:
# WN_MEAN_LOGIT, WN_STD_LOGIT: measured from 100 white-noise seeds passing
# through the ACTUAL pipeline (_modes_from_bytes → ROT_R.T @ (modes ROT_MU))
# followed by compute_phi_raw_ne with GLOBAL consecutive ratio search.
# Previous values (-0.579 / 0.550) were from synthetic sorted arrays and are
# invalidated by the global-search fix to compute_phi_raw_ne.
#
# GROUNDED_Z_NE = same z-offset (0.4σ) as EU GROUNDED_Z, now in NE logit-space.
# SEISMIC_Z_NE = same z-offset (-0.5σ) as EU FLAME boundary, in NE logit-space.
# FOAM_THRESHOLD_NE: logit(0.15) normalised through NE WN distribution.
# Full enwik9 empirical calibration pending (requires full 1GB enwik9 file).
#
# Geometry: log-multiplicative phi_prox (global ratio search) + logit z-score
# See: 6-Documentation/docs/NON_EUCLIDEAN_FLOAT.md, 6-Documentation/docs/EUCLIDEAN_ASSUMPTION_AUDIT.md
NE_CONSTANTS_CALIBRATED = True # pipeline-calibrated; update when enwik9 available
# ── White-noise logit distribution (100 seeds, actual pipeline rotation) ──────
# logit(phi_raw_ne) over white-noise chunks — used to normalise phi_z_ne
# NOTE: WN mean is positive (0.375) because ROT_R introduces PHI-ratio correlations
# even for white noise. Thresholds are offsets above/below THIS mean.
WN_MEAN_LOGIT = 0.375280410 # mean logit(phi_raw_ne) for white-noise input
WN_STD_LOGIT = 0.806610669 # std logit(phi_raw_ne) for white-noise input
# ── Phase thresholds (parallel to EU constants, in NE z-space) ────────────────
# In NE z-space, 0.0 = white noise mean. Offsets match the EU constants exactly.
GROUNDED_Z_NE = 0.4 # same offset as EU GROUNDED_Z (0.4σ above WN mean)
SEISMIC_Z_NE = -0.5 # same offset as EU SEISMIC/FLAME boundary (-0.5σ)
# logit(0.15) = -1.7346; normalised: (-1.735 - 0.375) / 0.807 = -2.616
FOAM_THRESHOLD_NE = -2.615737 # equiv. to foam_score >= 0.15 (was FOAM_THRESHOLD=0.15)
# ── Empirical calibration pending enwik9 availability ─────────────────────────────
# To calibrate with actual enwik9 data:
# Run: python tools/calibrate_geometry.py --enwik9 /path/to/enwik9 --output 5-Applications/scripts/soliton_constants_ne.py
# Current values are derived from EU constants with appropriate NE z-space offsets