dtypechar error in stats.norm.fit (Trac #43)

[scipy-gitbot]

Original ticket http://projects.scipy.org/scipy/ticket/43 on 2006-03-24 by trac user novin01 AT gmail.com, assigned to unknown.

Changing to .dtype.char fixes the problem, but I thought I'd post anyway.

This was on a P4/WinXP box with
python-2.4.2
numpy-0.9.4.win32-py2.4
scipy-0.4.4.win32-py2.4-P4SSE2

The following code replicates the error:

In [1]: from scipy import *
In [2]: x = stats.norm.rvs(size=1000,loc=0,scale=1)

In [3]: mu, sigma = stats.norm.fit(x)

exceptions.AttributeError Traceback (most recent call last)

\W2K4670\D$\GAS\Data\Processes

C:\Python24\lib\site-packages\scipy\stats\distributions.py in fit(self, data, _args, *_kwds)
726 # location and scale are at the end
727 x0 = args + (loc0, scale0)
--> 728 return optimize.fmin(self.nnlf,x0,args=(ravel(data),),disp=0)
729
730 def est_loc_scale(self, data, *args):

C:\Python24\lib\site-packages\scipy\optimize\optimize.py in fmin(func, x0, args, xtol, ftol, maxiter, maxfun, full_output, disp, retall)
189 sim = Num.zeros((N+1,),x0.dtypechar)
190 else:
--> 191 sim = Num.zeros((N+1,N),x0.dtypechar)
192 fsim = Num.zeros((N+1,),'d')
193 sim[0] = x0

AttributeError: 'numpy.ndarray' object has no attribute 'dtypechar'
In [4]: