Python中插值函数的优化

Question

I have a some model dimensions data (SangleRatio, RangleRatio, RudAngleRatio, RadRatio, SratioPole, RratioPole, Lwire, Nwire) and result from these data (nn) , and want to find the "optimal" model dimensions values (minimize "intnn" fcn):

import numpy as np
import scipy as sc
import scipy.interpolate as interpolate
import scipy.optimize    as optimize

exp= np.genfromtxt(ConsFile, delimiter="_")
SangleRatio, RangleRatio, RudAngleRatio, RadRatio, SratioPole, RratioPole, Lwire, Nwire = exp[:,0], exp[:,1], exp[:,2], exp[:,3], exp[:,4], exp[:,5], exp[:,6], exp[:,7]
nn = exp[:,8]

intnn = interpolate.Rbf(SangleRatio, RangleRatio, RudAngleRatio, RadRatio, SratioPole, RratioPole, Lwire, Nwire, nn,function='cubic')

initial_values = np.array([1.975, 1.525, 2.9, 3.6, 2.5, 5.335, 0.07, 22.25])
res=optimize.fmin_l_bfgs_b(intnn, x0=initial_values)

---

error messages:

Traceback (most recent call last):
  File "/home/peniak/pyth/optNew/interpolate10d_3.py", line 67, in <module>
    res=optimize.fmin_l_bfgs_b(intnn, x0=initial_values)
  File "/usr/local/lib/python2.7/dist-packages/scipy/optimize/lbfgsb.py", line 186, in fmin_l_bfgs_b
    **opts)
  File "/usr/local/lib/python2.7/dist-packages/scipy/optimize/lbfgsb.py", line 305, in _minimize_lbfgsb
    isave, dsave)
TypeError: failed to initialize intent(inout|inplace|cache) array -- input must be array but got <type 'numpy.float64'>

Answer 1

Problem solved...

def intNN(x):
p1, p2, p3, p4, p5, p6, p7, p8 = x
res=(-1)*intnn(p1, p2, p3, p4, p5, p6, p7, p8)
return float(res)

and then this function to optimize...