Matiq
25cf786c54
- Инфраструктура: RTTI-дампы (rtti_dsp/full.json), декомпиляции DSP-классов (decomp_*.txt), Ghidra-скрипты (Dump*.java, ImportRtti*.java, Diag.java), depthcurve/curve_fits LUT. - Поведенческая модель sim_v5.py + sim.py (RMSE ~0.3dB), утилиты (measure, tt_sweep, patchparam, sweep, harness_*, verify_sim). - summary.md + roadmap.md (контракт из manual M1-M12, топология пайплайна из OCR, фазы A-C). - pipeline_ocr.txt: OCR диаграммы Appendix A (mid/side ручка, trim/mix/bypass порядок).
52 lines
2.2 KiB
Python
52 lines
2.2 KiB
Python
#!/usr/bin/env python3
|
|
"""Fit parametric curves (v2): amount_total = A(depth)*S(sens)*H(sharp).
|
|
S normalized so S(12)=1, H(10)=1. A→1 as depth→large.
|
|
Saturation form: f(x)=1-exp(-(x/k)^m).
|
|
"""
|
|
import numpy as np
|
|
from scipy.optimize import curve_fit
|
|
|
|
def sat(x, k, m):
|
|
return 1 - np.exp(-(x / k) ** m)
|
|
|
|
def to_amount(red):
|
|
return 1 - 10 ** (-red / 20.0)
|
|
|
|
def fit(name, x, y, p0):
|
|
popt, _ = curve_fit(sat, x, y, p0=p0, maxfev=50000)
|
|
pred = sat(x, *popt)
|
|
print(f"{name}: k={popt[0]:.4f} m={popt[1]:.3f} RMSE={np.sqrt(np.mean((pred-y)**2)):.5f}")
|
|
for xi, yi, pi in zip(x, y, pred):
|
|
print(f" x={xi:6.2f} meas={yi:.4f} pred={pi:.4f}")
|
|
return popt
|
|
|
|
A_def = to_amount(7.70)
|
|
|
|
# A: depth sweep at S=H=1 (sens12 sharp10)
|
|
x = np.array([0.5, 0.864, 1.0, 2.0, 3.0, 5.0, 10.0, 20.0], float)
|
|
y = np.array([to_amount(d) for d in [7.14, 7.70, 7.91, 9.56, 11.29, 14.89, 24.25, 39.41]])
|
|
ka = fit('A(depth)', x, y, [3.0, 0.8])
|
|
|
|
# S: sens at depth0.864 (S saturates -> S(12)=1). amount = A_def * S
|
|
# S has a floor at sens=0 (still 2.89dB): S(x) = s0 + (1-s0)*sat(x)
|
|
def S(x, s0, k, m):
|
|
return s0 + (1 - s0) * sat(x, k, m)
|
|
x = np.array([0.0, 6.0, 12.0, 24.0, 48.0], float)
|
|
y = np.array([to_amount(d) for d in [2.89, 5.14, 7.70, 7.70, 7.70]]) / A_def
|
|
popt, _ = curve_fit(S, x, y, p0=[0.4, 3.0, 3.0], maxfev=50000)
|
|
pred = S(x, *popt)
|
|
print(f"S(sens): s0={popt[0]:.4f} k={popt[1]:.4f} m={popt[2]:.3f} RMSE={np.sqrt(np.mean((pred-y)**2)):.5f}")
|
|
for xi, yi, pi in zip(x, y, pred):
|
|
print(f" x={xi:6.2f} meas={yi:.4f} pred={pi:.4f}")
|
|
ks = popt
|
|
|
|
# H: sharp at depth0.864 -> H(10)=1
|
|
x = np.array([1.0, 3.0, 5.0, 10.0, 20.0], float)
|
|
y = np.array([to_amount(d) for d in [1.11, 3.71, 5.90, 7.70, 7.70]]) / A_def
|
|
kh = fit('H(sharp)/norm', x, y, [3.0, 1.5])
|
|
|
|
print("\ncross-check amount_total = A*S*H at default:")
|
|
print(f" A(0.864)*S(12)*H(10) = {sat(0.864,*ka):.4f}*1*1 -> red {(-20*np.log10(1-sat(0.864,*ka))):.2f} dB (meas 7.70)")
|
|
print(f" depth=10 sens=0: A(10)*S(0) = {sat(10,*ka):.4f}*{sat(0,*ks):.4f} = {sat(10,*ka)*sat(0,*ks):.4f} -> {( -20*np.log10(1-sat(10,*ka)*sat(0,*ks))):.2f} dB")
|
|
np.savez('/home/m/re-tools/curve_fits.npz',
|
|
k_depth=ka, k_sens=ks, k_sharp=kh, A_def=A_def) |