Metadata-Version: 2.4
Name: nomogeo
Version: 0.1.0
Summary: Exact observer-geometry kernel for visible precision and hidden-load calculus.
Author: J. R. Dunkley
License-Expression: BSD-3-Clause
Keywords: observer geometry,gaussian geometry,hidden load,visible precision,linear algebra
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Science/Research
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering :: Mathematics
Requires-Python: >=3.11
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy>=2.0
Requires-Dist: scipy>=1.15
Dynamic: license-file

# nomogeo

`nomogeo` is an exact observer-geometry kernel for visible precision, local visible calculus, hidden-load parametrisation, and quotient-side Gaussian contraction identities.

This PyPI package stages the `nomogeo` kernel only. The broader workspace also contains `nomodescent`, `evidence`, demonstrations, and research-facing materials, but those are intentionally excluded from this distribution.

## Install

```bash
python -m pip install nomogeo
```

## Dependencies

- `numpy>=2.0`
- `scipy>=1.15`

## Quick Start

```python
import numpy as np
from nomogeo import canonical_lift, hidden_load, inverse_visible_class, visible_precision

H = np.array([[3.0, 1.0], [1.0, 2.0]])
C = np.array([[1.0, 0.0]])

phi = visible_precision(H, C)
lift = canonical_lift(H, C)

T = np.diag([2.0, 1.0, 0.0])
Lambda = np.diag([0.3, 0.8])
X = inverse_visible_class(T, Lambda, lambda_representation="reduced")
load = hidden_load(T, X)
```

## Included Surface

- Exact visible precision `Phi_C(H) = (C H^{-1} C^T)^{-1}`
- Canonical lift and hidden projector
- Local visible calculus `(V, Q)` and determinant-curvature split
- Hidden-load parametrisation beneath a fixed ceiling
- Hidden-load transport and contraction
- Quotient-side Gaussian distances and contraction utilities

## Scope Boundaries

- The package is designed for exact finite-dimensional linear / Gaussian calculations.
- The fixed-ceiling inverse theorem requires the ceiling `T` as part of the input; it does not invert the global map `(H, C) -> Phi_C(H)`.
- For long hidden composition, use `hidden_contraction(...)` and `load_from_hidden_contraction(...)` rather than treating raw load coordinates as associative.
