mirror of
https://github.com/allaunthefox/Research-Stack.git
synced 2026-07-31 03:05:21 +00:00
0c9efac330
Squash the four overlapping feature branches into a single change set against main, eliminating cross-PR merge conflicts and the duplicated CI-fix scripts. What this brings in (merge order #79 -> #80 -> #81 -> #89): - #79 refactor(infra): shared utilities (4-Infrastructure/lib/*: q16, hashing, jsonl, fraction_utils) + the scripts/math-first/* validators that the math-check CI requires. - #80 feat(lean): Semantics.E8Sidon (1025 lines) -- Eisenstein coefficient identity E4^2 = E8 and the Sidon framework. E4_sq_eq_E8_coeff is fully proved (all Fourier-coefficient extraction machine-checked); the single residual gap is pinned to E4_sq_eq_E8_qExpansion (Mathlib lacks the valence formula / dim M8 = 1). 4 sorries + 1 axiom (e8_additive_completeness), all TODO(lean-port). - #81 refactor(lean): Float-free FixedPoint core (integer-only sqrt/log2/expNeg). E8Sidon.lean kept at #80's final 1025-line version (the #81 intermediate 438-line copy was overridden by merge order). - #89 feat(lean): Semantics.RRC.PolyFactorIdentity -- short-sleeve polynomial detection at the zerocopy limb boundary; now imports Semantics.E8Sidon for sigma3/sigma7/convolutionLHS (single source of truth) instead of inlining them. Conflict resolution: - flake.nix -> canonical rs-surface removal (Garnix shutdown). - scripts/math-first/* -> byte-identical across branches, clean. - .cursorrules / AGENTS.md -> unified; baselines + sorry inventory refreshed. Verification: - lake build (default aggregator): 3573 jobs, 0 errors. - lake build Semantics.RRC.PolyFactorIdentity (E8Sidon + FixedPoint + PolyFactor): 3655 jobs, 0 errors. Witnesses verified (sigma7 4 = 16513, convolutionLHS 6 = 2350). - Python tests: 68/68 pass. Note: the "Workers Builds: researchstack" check is a preexisting external Cloudflare build unrelated to this change (no branch touches 4-Infrastructure/cloudflare/). Build: 3573 jobs (default), 3655 jobs (narrow), 0 errors Co-Authored-By: Allaun Silverfox <bigdataiscoming+9i37y6j2@protonmail.com>
113 lines
2.9 KiB
Python
113 lines
2.9 KiB
Python
import numpy as np
|
|
import sys
|
|
from pathlib import Path
|
|
|
|
sys.path.insert(0, str(Path(__file__).resolve().parents[3] / "4-Infrastructure"))
|
|
from lib.q16 import Q16_ONE, q16_mul
|
|
|
|
|
|
# Phase 1 — Build BurgersTriadCore
|
|
# Q16.16 in the AVM hot path
|
|
|
|
# Q16.16 Constants
|
|
Q16_SHIFT = 16
|
|
Q16_MAX = (1 << 31) - 1
|
|
Q16_MIN = -(1 << 31)
|
|
|
|
def float_to_q16(f: float) -> int:
|
|
return q16_sat(int(f * Q16_ONE))
|
|
|
|
def q16_to_float(q: int) -> float:
|
|
return q / Q16_ONE
|
|
|
|
_sat_count = 0
|
|
|
|
def get_sat_count() -> int:
|
|
global _sat_count
|
|
return _sat_count
|
|
|
|
def reset_sat_count():
|
|
global _sat_count
|
|
_sat_count = 0
|
|
|
|
def q16_sat(x: int) -> int:
|
|
"""Saturate to 32-bit signed integer."""
|
|
global _sat_count
|
|
if x > Q16_MAX:
|
|
_sat_count += 1
|
|
return Q16_MAX
|
|
if x < Q16_MIN:
|
|
_sat_count += 1
|
|
return Q16_MIN
|
|
return x
|
|
def triad_rhs(a: tuple[int, int, int], nu_eff: int) -> tuple[int, int, int]:
|
|
"""
|
|
Triad equations (Burgers):
|
|
da1/dt = -nu_eff a1 + 1/2(a1a2 + a2a3)
|
|
da2/dt = -4nu_eff a2 - 1/2 a1^2 + a1a3
|
|
da3/dt = -9nu_eff a3 - 3/2 a1a2
|
|
"""
|
|
a1, a2, a3 = a
|
|
|
|
# Precompute products
|
|
a1_a2 = q16_mul(a1, a2)
|
|
a2_a3 = q16_mul(a2, a3)
|
|
a1_a3 = q16_mul(a1, a3)
|
|
a1_a1 = q16_mul(a1, a1)
|
|
|
|
# 1/2 is (1 << 15), 3/2 is (3 << 15), etc. Or just multiply and divide by 2
|
|
# To maintain Q16 semantics, we can multiply by Q16 constants:
|
|
HALF = 1 << 15
|
|
THREE_HALVES = 3 << 15
|
|
|
|
# da1/dt = -nu_eff * a1 + 1/2 * (a1a2 + a2a3)
|
|
t1_1 = -q16_mul(nu_eff, a1)
|
|
t1_2 = q16_mul(HALF, q16_sat(a1_a2 + a2_a3))
|
|
da1 = q16_sat(t1_1 + t1_2)
|
|
|
|
# da2/dt = -4 * nu_eff * a2 - 1/2 * a1^2 + a1a3
|
|
FOUR = 4 << Q16_SHIFT
|
|
t2_1 = -q16_mul(q16_mul(FOUR, nu_eff), a2)
|
|
t2_2 = -q16_mul(HALF, a1_a1)
|
|
t2_3 = a1_a3
|
|
da2 = q16_sat(q16_sat(t2_1 + t2_2) + t2_3)
|
|
|
|
# da3/dt = -9 * nu_eff * a3 - 3/2 * a1a2
|
|
NINE = 9 << Q16_SHIFT
|
|
t3_1 = -q16_mul(q16_mul(NINE, nu_eff), a3)
|
|
t3_2 = -q16_mul(THREE_HALVES, a1_a2)
|
|
da3 = q16_sat(t3_1 + t3_2)
|
|
|
|
return (da1, da2, da3)
|
|
|
|
def rk2_step(a: tuple[int, int, int], nu_eff: int, dt: int) -> tuple[int, int, int]:
|
|
"""Midpoint RK2 step in Q16.16."""
|
|
a1, a2, a3 = a
|
|
|
|
k1 = triad_rhs(a, nu_eff)
|
|
|
|
# Midpoint
|
|
half_dt = dt >> 1
|
|
a_mid = (
|
|
q16_sat(a1 + q16_mul(k1[0], half_dt)),
|
|
q16_sat(a2 + q16_mul(k1[1], half_dt)),
|
|
q16_sat(a3 + q16_mul(k1[2], half_dt))
|
|
)
|
|
|
|
k2 = triad_rhs(a_mid, nu_eff)
|
|
|
|
# Full step
|
|
a_next = (
|
|
q16_sat(a1 + q16_mul(k2[0], dt)),
|
|
q16_sat(a2 + q16_mul(k2[1], dt)),
|
|
q16_sat(a3 + q16_mul(k2[2], dt))
|
|
)
|
|
|
|
return a_next
|
|
|
|
def energy3(a: tuple[int, int, int]) -> int:
|
|
"""E = 1/2 (a1^2 + a2^2 + a3^2) in Q16.16."""
|
|
a1, a2, a3 = a
|
|
sum_sq = q16_sat(q16_sat(q16_mul(a1, a1) + q16_mul(a2, a2)) + q16_mul(a3, a3))
|
|
HALF = 1 << 15
|
|
return q16_mul(HALF, sum_sq)
|