Class fit_function

source code

object --+
         |
        fit_function

Known Subclasses:

• fit_cubic
• , fit_diff_of_exp
• , fit_exponential
• , fit_linear
• , fit_quadratic
• , fit_quadratic_at_vertex

Abstract super-class for fitting explicit functions to 1D arrays of data
using least squares.

xs -- independent variable data
ys -- dependent variable data
pars_ic -- initial values defining the function

Optional algorithmic parameters to minpack.leastsq can be passed in the
algpars argument: e.g.,
ftol -- Relative error desired in the sum of squares (default 1e-6).
xtol -- Relative error desired in the approximate solution (default 1e-6).
gtol -- Orthogonality desired between the function vector
        and the columns of the Jacobian (default 1e-8).

Other parameters may be used for concrete sub-classes. Pass these as a dict
or args object in the opts argument.

Returns an args object with attributes:

ys_fit --   the fitted y values corresponding to the given x data,
pars_fit -- the function parameters at the fit
info --     diagnostic feedback from the leastsq algorithm
results --  dictionary of other function specific information (such as peak
             position)