slug: scipy-optimize-curve*fit date*published: 2018-03-15T07:23:49 date_updated: 2018-03-15T07:23:49 tags: Acedemic Notes, English Posts excerpt: "the gist of it is, when fitting, use p0 argument to give initial point for fitting." –-

Isn't able to estimate the covariance of the parameter, thus raising the OptimizeWarning below Scipy Curve_fit

Today's lab ran into a problem that when people try to fit a function with

exponential and square root in it, the results is either off, or the

prameter does not work.

The problem is that if the fitting function tries to go through `undefined`

points in your function, it throws out the warning above.

The importance of guessing is discussed in this thread, but the gist of it is, when fitting, use `p0`

argument to give initial point for fitting.

popt, pcov = curve_fit(func2, np.arange(10), np.arange(10), p0=(0, 0))