Datasets:
Formats:
csv
Languages:
English
Size:
< 1K
Tags:
mathematics
information-theory
quantum-information
representation-theory
bosonic-memory
task-relative-compression
Download code/demo.py from PureOne/ORYNTHRA-H6: direct link, hf CLI and curl.
- Browser
- Download file 1.84 kB
-
https://huggingface.co/datasets/PureOne/ORYNTHRA-H6/resolve/main/code/demo.py
- Command line
-
hf download hf://datasets/PureOne/ORYNTHRA-H6/code/demo.py
-
curl -L -o demo.py https://huggingface.co/datasets/PureOne/ORYNTHRA-H6/resolve/main/code/demo.py
1.84 kB
| """Small executable example; no optical hardware or external data is needed.""" | |
| from pathlib import Path | |
| import numpy as np | |
| from orynthra import (LinearEdge, observable_closure, six_wing_actions, | |
| storage_cutoff, symmetric_cutoff, simplex_code, | |
| maximum_packed_dimension) | |
| def main() -> None: | |
| actions = six_wing_actions(36) | |
| observation = np.eye(36, dtype=int)[0:1] | |
| five, _ = observable_closure([observation], [LinearEdge(0,0,a) for a in actions[:5]]) | |
| six, _ = observable_closure([observation], [LinearEdge(0,0,a) for a in actions]) | |
| x = np.zeros(36, dtype=int) | |
| y = x.copy() | |
| y[6] = 1 | |
| assert np.array_equal(five[0] @ x, five[0] @ y) | |
| for wing in [5,4,3,2,1,0]: | |
| x, y = actions[wing] @ x, actions[wing] @ y | |
| assert x[0] != y[0] | |
| print("ORYNTHRA-H6: three exact finite demonstrations") | |
| print(f"36 raw binary coordinates -> {len(five[0])} task coordinates without Kaya.") | |
| print(f"All six branch generators -> {len(six[0])} task coordinates.") | |
| print("A hidden seventh coordinate becomes visible under Kaya,Evie,Enya,Raven,Tachy,Lily.") | |
| print("The task compression intentionally discards unprotected raw distinctions.") | |
| print() | |
| for messages in [6,64,256,4096]: | |
| print(f"M={messages}: two-mode storage cutoff={storage_cutoff(messages,2)}, " | |
| f"full-permutation coherent cutoff={symmetric_cutoff(messages,2)}") | |
| v, number, cutoff = simplex_code(6,2) | |
| print(f"H6 Gram residual: {np.linalg.norm(v.T @ v-np.eye(6)):.3e}") | |
| print(f"H6 uniform mean photon number: {np.trace(v.T @ number @ v)/6:.12g}") | |
| matched = maximum_packed_dimension([1,1,4,4],[1,2,3,4]) | |
| print(f"Multiplicity-trap matched dimension: {matched}/10; exact mean defect={2*(1-matched/10):.6g}") | |
| if __name__ == "__main__": | |
| main() | |