Reputation: 3591
I tried running .fit
on a dweibull
distribution from scipy.stats
I get a list of three values but what are they? looking here: Scipy Documentation
I can't see any description of what fit returns.
Upvotes: 3
Views: 2374
Reputation: 26030
It's a tuple of <shapes>, loc, scale
. dweibull has one shape parameter, so you get three items. norm
does not have shape parameters, hence you only get two for loc and scale, etc
The idiom is that you can unpack the output of fit
into a call to pdf
, cdf
et al:
>>> from scipy import stats
>>> data = stats.beta.rvs(a=1, b=2, size=100, random_state=101)
>>> xxx = stats.beta.fit(data)
>>> stats.beta.pdf(0.1, *xxx)
1.7748574630838663
>>> stats.beta.mean(*xxx)
0.33473342172664911
Upvotes: 4