fused_chaos_index.tier2¶
Tier-2 advanced analyses for multi-artifact comparisons and discoveries.
Overview¶
Tier-2 commands provide sophisticated analysis workflows:
- universality_sweep - Compare quantum-mass distributions across multiple runs
- fingerprint - Cosmic vs. Collatz distribution fingerprinting with correlation analysis
- collatz_summary - Comprehensive statistics for Collatz runs (manuscript-ready)
- stopping_time - Investigate stopping time vs. quantum mass correlations
- plot_16m_discovery - Visualize 16M Collatz discovery (optional PNG if matplotlib available)
Use Cases¶
Universality Analysis¶
Compare multiple quantum-mass distributions to test universality hypothesis:
from fused_chaos_index.tier2 import run_universality_sweep
# Compare multiple runs
result = run_universality_sweep(
inputs=["run1.npz", "run2.npz", "run3.npz"],
output_dir="./results",
threshold=5e-7,
seed=42,
)
print(f"Mean quantum mass: {result['mean_quantum_mass']:.6e}")
print(f"Std dev: {result['std_quantum_mass']:.6e}")
Cosmic vs. Collatz Fingerprinting¶
Compare distribution fingerprints between cosmic data and Collatz sequences:
from fused_chaos_index.tier2 import run_cosmic_vs_collatz_fingerprint
# Fingerprint comparison
result = run_cosmic_vs_collatz_fingerprint(
collatz_npz="collatz_run.npz",
smacs_npz="smacs_catalog.npz",
output_dir="./results",
k=10,
n_modes=10,
seed=42,
)
print(f"Mass correlation: {result['mass_kappa_correlation']:.3f}")
Collatz Run Summary¶
Generate comprehensive statistics for Collatz runs:
from fused_chaos_index.tier2 import run_collatz_run_summary
# Manuscript-ready summary
result = run_collatz_run_summary(
collatz_npz="collatz_run.npz",
baseline_npz="baseline_run.npz",
output_dir="./results",
runtime_seconds=136.5,
)
print(f"Mean quantum mass: {result['mass_mean']:.6e}")
print(f"Spectral gap ratio: {result['gap_ratio']:.3f}")
Stopping Time Analysis¶
Test for correlations between Collatz stopping time and quantum mass:
from fused_chaos_index.tier2 import run_stopping_time_vs_mass
# Association test
result = run_stopping_time_vs_mass(
input_npz="collatz_run.npz",
output_dir="./results",
sample_size=200000,
max_steps=5000,
seed=42,
)
print(f"Correlation: {result['correlation']:.3f}")
print(f"P-value: {result['p_value']:.6f}")
Data Requirements¶
Tier-2 commands expect NPZ files with specific fields:
- quantum_mass (preferred) or mass or M - Mass/eigenvalue array
- positions (N×D) or ra+dec - Spatial coordinates
- kappa (optional) - Convergence field for lensing catalogs
- eigenvalues (optional) - For spectral gap analysis
See NPZ contracts for detailed specifications.
API Reference¶
fused_chaos_index.tier2
¶
collatz_stopping_time_uint64(n0, *, max_steps)
¶
Compute Collatz stopping time for many starting values.
Returns (stop_time, status): - stop_time: int32, 0..max_steps, max_steps means censored, -1 means overflow - status: int8 codes: 0=reached 1, 1=censored, 2=overflow
run_collatz_run_summary(*, output_dir, collatz_npz, baseline_npz=None, threshold=5e-07, cosmic_target=68.0, runtime_seconds=None)
¶
Summarize a Collatz run artifact (NPZ) into manuscript-ready stats.
Offline-first: consumes local NPZ files only and writes a JSON manifest + NPZ results.
Required inputs:
- collatz_npz: must contain quantum_mass (or mass/M)
Optional inputs: - baseline_npz: second NPZ for comparison (e.g., 10M baseline)
Notes:
- If dark_percent exists in NPZ, it is recorded.
- If not, dark% is computed from the mass vector and threshold.
- If eigenvalues exists, the spectral gap and gap ratio are computed.
- Output NPZ uses flattened dict format (e.g., mass_stats__n) to avoid
dtype=object, allowing secure loading with allow_pickle=False.
run_cosmic_vs_collatz_fingerprint(*, output_dir, collatz_npz, smacs_npz, flamingo_npz=None, threshold=5e-07, k=10, n_modes=10)
¶
Tier-2 Path 2: cross-domain fingerprint comparison.
Inputs are purely local artifacts (NPZ files). This runner computes:
- distribution fingerprints of quantum_mass and (for SMACS) kappa
- correlations between SMACS quantum mass and kappa
- optional Collatz low-energy spectrum summaries (if eigenvalues included)
It does not claim physical causality; it only writes descriptive statistics.
Output NPZ uses flattened dict format (e.g., collatz_fingerprint__n) to avoid dtype=object, allowing secure loading with allow_pickle=False.
run_plot_16m_discovery(*, output_dir, collatz_npz, threshold=5e-07, cosmic_target=68.0)
¶
Generate a lightweight plot for the 16M-style Collatz discovery artifact.
Offline-first and skip-safe: - reads one local NPZ - attempts to write a PNG if matplotlib is available - always writes a results NPZ + JSON manifest
This is intentionally minimal (no seaborn) to keep base installs light.
Output NPZ avoids dtype=object (issues saved as JSON string) to allow secure loading with allow_pickle=False.
run_stopping_time_vs_mass(*, output_dir, input_npz, sample_size=200000, max_steps=5000, seed=42, bootstrap=False, bootstrap_iters=200)
¶
Tier-2 Path 3: stopping time vs quantum-mass association test.
Offline-first: reads one local NPZ, writes results NPZ + JSON manifest.
run_universality_sweep(*, output_dir, inputs, threshold=5e-07, plot=False)
¶
Tier-2 Path 1: compare multiple quantum-mass artifacts under a shared threshold.
This is offline-first: it only reads local NPZ inputs and writes local outputs.
Each input NPZ must contain one of the keys: - quantum_mass (preferred) - mass - M
Optional keys used if present: - dark_percent (original run's reported percentage) - k_modes or eigenvalues (used to infer k)