Tukukkk

I want to fit a set of data with a simple sin^2 function and want to determine its minima based on the fitted parameters.

Here's my code:

import numpy as np

import matplotlib.pyplot as plt

from scipy import optimize



data = np.loadtxt('data.txt', usecols=(0,1))

x = data[:,0]*np.pi/180

y = data[:,1]



plt.scatter(x, y, c='red')



def sine(t,a,b,c):

    return a*(np.sin(b*(t-c)))**2



params, cov = optimize.curve_fit(sine, x, y, p0=[9500, 0.5, 0])

print(params)



t = np.linspace(0, 2*np.pi/3, 120) 

plt.plot(t, sine(t, *params), 'black')



plt.show()



optimize.minimize(sine(t, *params), x0=0)

Everything is fine except for the minimize call as I get the following error (with a full traceback):

TypeError                                 Traceback (most recent call last)

~DocumentsCNRCalibrazione_lamine_20181112Fit.py in <module>()

     23 plt.show()

     24 

---> 25 optimize.minimize(sine(t, *params), x0=0)



~Anaconda3libsite-packagesscipyoptimize_minimize.py in minimize(fun, x0, args, method, jac, hess, hessp, bounds, constraints, tol, callback, options)

    442         return _minimize_cg(fun, x0, args, jac, callback, **options)

    443     elif meth == 'bfgs':

--> 444         return _minimize_bfgs(fun, x0, args, jac, callback, **options)

    445     elif meth == 'newton-cg':

    446         return _minimize_newtoncg(fun, x0, args, jac, hess, hessp, callback,



~Anaconda3libsite-packagesscipyoptimizeoptimize.py in _minimize_bfgs(fun, x0, args, jac, callback, gtol, norm, eps, maxiter, disp, return_all, **unknown_options)

    911     else:

    912         grad_calls, myfprime = wrap_function(fprime, args)

--> 913     gfk = myfprime(x0)

    914     k = 0

    915     N = len(x0)



~Anaconda3libsite-packagesscipyoptimizeoptimize.py in function_wrapper(*wrapper_args)

    290     def function_wrapper(*wrapper_args):

    291         ncalls[0] += 1

--> 292         return function(*(wrapper_args + args))

    293 

    294     return ncalls, function_wrapper



~Anaconda3libsite-packagesscipyoptimizeoptimize.py in approx_fprime(xk, f, epsilon, *args)

    686 

    687     """

--> 688     return _approx_fprime_helper(xk, f, epsilon, args=args)

    689 

    690 



~Anaconda3libsite-packagesscipyoptimizeoptimize.py in _approx_fprime_helper(xk, f, epsilon, args, f0)

    620     """

    621     if f0 is None:

--> 622         f0 = f(*((xk,) + args))

    623     grad = numpy.zeros((len(xk),), float)

    624     ei = numpy.zeros((len(xk),), float)



~Anaconda3libsite-packagesscipyoptimizeoptimize.py in function_wrapper(*wrapper_args)

    290     def function_wrapper(*wrapper_args):

    291         ncalls[0] += 1

--> 292         return function(*(wrapper_args + args))

    293 

    294     return ncalls, function_wrapper



TypeError: 'numpy.ndarray' object is not callable.

I'm missing something but I don't know what.

I'm adding the data file to make this program run, as suggested

0   405

5   20

10  350

15  1380

20  2900

25  4750

30  6450

35  8100

40  9100

45  9800

50  10100

55  10250

60  9400

65  8400

70  6430

75  4900

80  3030

85  1500

90  400

95  17

100 410

105 1550

110 3100

115 4850

120 6780

minimize expects a function as first argument, however, you currently pass

sine(t, *params)

which is a numpy array.

You can fix this and do:

print(optimize.minimize(sine, x0=[0], args=tuple(params)))

This will print

      fun: 2.4080485986582715e-12

 hess_inv: array([[1.15258817e-05]])

      jac: array([8.19961349e-09])

  message: 'Optimization terminated successfully.'

     nfev: 18

      nit: 4

     njev: 6

   status: 0

  success: True

        x: array([0.09203053])

In the documentation of scipy, optimize.minimize function takes ndarray or shape(n) as an input for x, not an integer. I think the error is raised from there because in their error trace

--> 913     gfk = myfprime(x0)

the error is raised form this function.

Documentation link.