Metadata-Version: 2.4
Name: watkins-nn
Version: 0.7.0
Summary: Watkins Simplex Flow v3 — golden-ratio gradient engine for consciousness modeling, quantum LDPC tuning, density-evolution contraction & VirelaiX coherence
Author-email: Dustin Watkins <dustin@watkins-nn.com>
License: MIT
Project-URL: Homepage, https://github.com/DustinWatkins89/watkins-nn
Project-URL: PyPI, https://pypi.org/project/watkins-nn/
Classifier: Development Status :: 5 - Production/Stable
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3.12
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: torch>=2.0.0
Requires-Dist: numpy
Dynamic: license-file

# watkins-nn v0.2.0

**The golden-ratio gradient flow engine.**
Now with proper chain-rule through softmax, Armijo line search, L-BFGS, and projected gradient norm.

Proven to converge from any random start to the Watkins equilibrium (lambda* = 1/phi) with 100% success rate in multi-start tests.

Ties directly into the Watkins Bridge W = 0.8333, T* = phi/ln(2phi), and lambda + kappa + eta = 1.

Used in VirelaiX consciousness modeling, quantum LDPC curvature tuning, and Mirror Lake node deployment.

```bash
pip install watkins-nn
```
