fit multiple parametric curves with scipy

Question

I have a set (at least 3) of curves (xy-data). For each curve the parameters E and T are constant but different. I\'m searching the coefficients a,n and m for the best fit o

Answer 1

One way to do this is use scipy.optimize.leastsq instead (curve_fit is a convenience wrapper around leastsq).

Stack the x data in one dimension; ditto for the y data. The lengths of the 3 individual datasets don't even matter; let's call them n1, n2 and n3, so your new x and y will have a shape (n1+n2+n3,).

Inside the function to optimize, you can split up the data at your convenience. It will not be the nicest function, but this could work:

def function(x, E, T, a, n, m):
    return x/E + (a/n+1)*T^(n+1)*x^m

def leastsq_function(params, *args):
    a = params[0]
    n = params[1]
    m = params[2]
    x = args[0]
    y = args[1]
    E = args[2]
    T = args[3]
    n1, n2 = args[2]

    yfit = np.empty(x.shape)
    yfit[:n1] = function(x[:n1], E[0], T[0], a, n, m)
    yfit[n1:n2] = function(x[n1:n2], E[1], T[1], a, n, m)
    yfit[n2:] = function(x[n2:], E[2], T[2], a, n, m)

    return y - yfit


params0 = [a0, n0, m0]
args = (x, y, (E0, E1, E2), (T0, T1, T2), (n1, n1+n2))
result = scipy.optimize.leastsq(leastsq_function, params0, args=args)

I have not tested this, but this is the principle. You're now splitting up the data into 3 different calls inside the function that is to be optimized.

Note that scipy.optimize.leastsq simply requires a function that returns whatever value you'd like to be minized, in this case the difference between your actual y data and the fitted function data. The actual important variables in leastsq are the parameters you want to fit for, not the x and y data. The latter are passed as extra arguments, together with the sizes of three separate datasets (I'm not using n3, and I've done some juggling with the n1+n2 for convenience; keep in mind that the n1 and n2 inside leastsq_function are local variables, not the original ones).

Since this is an awkward function to fit (it probably won't have a smooth derivative, for example), it is quite essential to

• provide good starting values (params0, so all the ...0 values).

• don't have data or parameters which span orders of magnitude. The closer everything is around 1 (a few orders of magnitude is certainly ok), the better.

Answer 2

thanks Evert for the reply.

Exactly what I needed to know!!

I have simplyfied the function as far as possible, as you suggested. However, the task was to find ONE set of A,m,n to fit all curves. So my code look like this:

import numpy
import math
from scipy.optimize import leastsq

#+++++++++++++++++++++++++++++++++++++++++++++
def fit(x,T,A,n,m):
  return A/(n+1.0)*math.pow(T,(n+1.0))*numpy.power(x,m)
#+++++++++++++++++++++++++++++++++++++++++++++
def leastsq_func(params, *args):
  cc=args[0]   #number of curves
  incs=args[1] #number of points 
  x=args[2]
  y=args[3]
  T=args[4:]

  A=params[0]
  n=params[1]
  m=params[2]

  yfit=numpy.empty(x.shape)
  for i in range(cc):
    v=i*incs
    b=(i+1)*incs
    if b<cc:
     yfit[v:b]=fit(x[v:b],T[i],A,n,m)
    else:
     yfit[v:]=fit(x[v:],T[i],A,n,m)

  return y-yfit
#+++++++++++++++++++++++++++++++++++++++++++++
Ts  =[10,100,1000,10000]    #4 T-values for 4 curves
incs=10                     #10 datapoints in each curve
x=["measured data"]   #all 40 x-values
y=["measrued data"]   #all 40 y-values
x=numpy.array(x)   
y=numpy.array(y)   

params0=[0.001,1.01,-0.8]   #parameter guess

args=[len(Ts),incs,x,y]
for c in Ts:
  args.append(c) 
args=tuple(args)   #doesn't work if args is a list!!

result=leastsq(leastsq_func, params0, args=args)

Works like clockwork.

At first I put the Ts in the params0 list and they were modified during iteration leading to nonsense results. Obvious, if you think about it. Afterwards ;-)

So, Vielen Dank! J.