Download REPRODUCIBILITY.md from PureOne/AEQORIA-Sharp-Holonomy-Observation: direct link, hf CLI and curl.
- Browser
- Download file 3.37 kB
-
https://huggingface.co/datasets/PureOne/AEQORIA-Sharp-Holonomy-Observation/resolve/main/REPRODUCIBILITY.md
- Command line
-
hf download hf://datasets/PureOne/AEQORIA-Sharp-Holonomy-Observation/REPRODUCIBILITY.md
-
curl -L -o REPRODUCIBILITY.md https://huggingface.co/datasets/PureOne/AEQORIA-Sharp-Holonomy-Observation/resolve/main/REPRODUCIBILITY.md
Reproducing ÆQORIA
This Hugging Face edition preserves the original research release under
research/ and as releases/AEQORIA_Complete_Research_Release.zip.
The PDF, mathematical source, reference code, and original test record are
unchanged. Packaging validation is recorded separately under validation/.
Environment
The original run used Python 3.13.5, NumPy 2.3.5, SciPy 1.17.0, SymPy 1.14.0, and mpmath 1.3.0. Use Python 3.13 with the pinned requirements for the closest reproduction. The uploader has its own environment and only needs Python 3.10 or newer; installing the uploader does not install scientific dependencies.
From a downloaded copy of this repository:
cd research
python -m venv .venv-research
Activate the environment on Windows PowerShell with
.venv-research\Scripts\Activate.ps1, on Windows Command Prompt with
.venv-research\Scripts\activate.bat, or on macOS/Linux with
source .venv-research/bin/activate. Then:
python -m pip install -r requirements.txt
python verify_aeqoria.py --output ../local_results/verification.json
The expected summary is:
PASS: 142 checks; categories: {'symbolic': 10, 'algebra': 7, 'numerical': 116, 'high_precision': 9}
The command also writes asymptotics.json beside its selected output. Using
local_results/ keeps the original evidence record unchanged. Runtime,
interpreter version, and some floating-point residuals may vary. The verifier's
embedded date is a release label; the separate packaging report identifies
the fresh execution.
Regenerate the example tables
python create_examples.py
This evaluates the explicit circle design with 80-digit mpmath arithmetic and
updates research/results/circle_examples.csv and .json. Both files were
regenerated in a separate temporary copy during packaging and matched the
originals byte for byte. These values describe a constructive circle family,
not certified finite-length global optima.
Verify original release integrity
Run the following from research/:
from pathlib import Path
import hashlib
root = Path('.')
for line in (root / 'MANIFEST_SHA256.txt').read_text().splitlines():
expected, name = line.split(' ', 1)
assert hashlib.sha256((root / name).read_bytes()).hexdigest() == expected, name
print('All original release hashes match.')
The original manifest covers 38 files. The enclosing original ZIP has SHA-256:
c29d3d15986dd1655e3758c2cf377582e661d7a687de1ff4e275f14a0db90b6b
Rebuild the manuscript
Install a TeX distribution with pdfLaTeX, latexmk, and the packages listed in
the main source. From research/:
latexmk -pdf -interaction=nonstopmode AEQORIA_Sharp_Holonomy_Observation.tex
The supplied PDF has 67 physical pages. The original release records its PDF build and layout inspection; packaging checks the supplied PDF page count and integrity, and does not claim a fresh LaTeX rebuild.
Evidence boundary
The suite checks selected symbolic identities, finite-dimensional algebra,
numerical instances, and high-precision asymptotics. It does not exhaust all
loops, formally verify the manuscript, establish worldwide priority, or replace
independent mathematical review. Refer to research/CLAIM_LEDGER.md for the
proof/evidence map and to research/STATUS.json for the original status.