def doFit(self):
guess = []
scale = [] # scaling factor of variables
ub = [] # upper bound
lb = [] # lower bound
for k in self.fitParam:
v = self.baseParam[k]
if isinstance(v, tuple):
vmin = v[1]
vmax = v[2]
invS = abs(vmax - vmin)
v = v[0]
else:
vmin = -1e100
vmax = 1e100
invS = min(1e5, max(1e-6, abs(float(v))))
guess.append(v)
lb.append(vmin)
ub.append(vmax)
scale.append(1. / invS)
#result, success = leastsq(self.fun, guess, (), self.jac, warning=True, factor=1., diag=scale)
#result, cov_x, infodict, mesg, success = leastsq(self.fun, guess, (), self.jac, warning=True, factor=1., full_output=1)
#print '*****', success, infodict['nfev'], mesg
LSP = NLLSP(self.fun,
guess,
df=self.jac,
ub=ub,
lb=lb,
scale=scale,
xtol=1e-12,
ftol=1e-12,
gtol=1e-12,
maxFunEvals=100)
r = LSP.solve(self.solver)
result = r.xf
if isinstance(result, np.float):
result = [result]
e = self.fun(result)
param = {}
for i, k in enumerate(self.fitParam):
param[k] = result[i]
return param, norm(e)
def doFit(self):
guess=[]
scale=[] # scaling factor of variables
ub=[] # upper bound
lb=[] # lower bound
for k in self.fitParam:
v = self.baseParam[k]
if isinstance(v, tuple):
vmin = v[1]
vmax = v[2]
invS = abs(vmax-vmin)
v=v[0]
else:
vmin = -1e100
vmax = 1e100
invS = min(1e5, max(1e-6, abs(float(v))))
guess.append(v)
lb.append(vmin)
ub.append(vmax)
scale.append(1./invS)
#result, success = leastsq(self.fun, guess, (), self.jac, warning=True, factor=1., diag=scale)
#result, cov_x, infodict, mesg, success = leastsq(self.fun, guess, (), self.jac, warning=True, factor=1., full_output=1)
#print '*****', success, infodict['nfev'], mesg
LSP = NLLSP(self.fun, guess, df = self.jac, ub=ub, lb=lb, scale=scale, xtol = 1e-12, ftol = 1e-12, gtol=1e-12, maxFunEvals = 100)
r = LSP.solve(self.solver)
result = r.xf
if isinstance(result, np.float):
result = [result]
e = self.fun(result)
param = {}
for i,k in enumerate(self.fitParam):
param[k] = result[i]
return param, norm(e)
startPoint['LifeTime'] = 1.0
startPoint['LifeTime'] = 1.0
startPoint['ScatterAmp'] = 1.0
p = NLLSP(objfunc, startPoint)
# p.constraints = [
# (2*c+a-10)**2 < 1.5 + 0.1*b, (a-10)**2<1.5,
# a[0]>8.9, a+b > [ 7.97999836, 7.8552538 ],
# a < 9,
# (c-2)**2 < 1,
# b < -1.02,
# c > 1.01,
# (b + c * log10(a).sum() - 1) ** 2==0
# ]
r = p.solve('ralg')
RAmp, RLifeTime, ROffset, RScatterAmp = r(Amp, LifeTime, Offset, ScatterAmp)
print(RAmp, RLifeTime, ROffset, RScatterAmp)
# ------------------------------------------------
Plot = False
if Plot:
from Display import ForkDisplay
print "\nPlotting: Aligned Data:"
ForkDisplay(Time,
AlignedRaw,
Title="Aligned Raw Data",
YAxis="Intensity (Counts)")
#optional: gradient
def df(x):
r = zeros((3,2))
r[0,0] = 2*x[0]
r[0,1] = 2*x[1]
r[1,0] = 4*x[0]**3
r[1,1] = 4*x[1]**3
r[2,0] = 1
return r
# init esimation of solution - sometimes rather pricise one is very important
x0 = [1.5, 8]
#p = NLLSP(f, x0, diffInt = 1.5e-8, xtol = 1.5e-8, ftol = 1.5e-8)
# or
# p = NLLSP(f, x0)
# or
p = NLLSP(f, x0, df = df, xtol = 1.5e-8, ftol = 1.5e-8)
#optional: user-supplied gradient check:
p.checkdf()
#r = p.solve('scipy_leastsq', plot=1, iprint = -1)
#or using converter lsp2nlp:
r = p.solve('nlp:ralg', iprint = 1, plot=1)
#r = p.solve('nlp:ipopt',plot=1), r = p.solve('nlp:algencan'), r = p.solve('nlp:ralg'), etc
#(some NLP solvers require additional installation)
print 'x_opt:', r.xf # 2.74930862, +/-2.5597651
print 'funcs Values:', p.f(r.xf) # [-0.888904734668, 0.0678251418575, -0.750691380965]
print 'f_opt:', r.ff, '; sum of squares (should be same value):', (p.f(r.xf) ** 2).sum() # 1.35828942657