Compute derivatives by default numerical differentiation for all non-frozen parameters, This must be overridden for fits which have analytic derivatives. See example in class documentation. Note in particular that the array returned is organized with all the values for a point in each row. The columns correspond to the parameters. Reimplemented from fit. Definition at line 877 of file fitting_toolkit.py. 00877 : 00878 #analytic derivatives for a 1-d gaussian 00879 #z0+a*exp( -(x-xmu)**2/(2*xsig**2) ) 00880 z0, a, xmu, xsigma = self.funcparams 00881 n=self.pointcount 00882 x=self.xarray[0,:n] 00883 xsigi=-1.0/(2.0*xsigma**2) 00884 dx=x-xmu 00885 dx2=dx*dx 00886 expfact=Numeric.exp(xsigi*dx2) 00887 z=a*expfact 00888 00889 dd = zeros((n, 4), self.atype) 00890 dd[:,0]=1.0 00891 dd[:,1]=expfact 00892 dd[:,2]=(-2.0*xsigi)*(dx*z) 00893 dd[:,3]=(-2.0*xsigi/xsigma)*(dx2*z) 00894 00895 return dd 00896
|
1.5.4