You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have attached the script test_minuit.py and the output test_minuit_output.txt
I get that shows the incorrect results.
What version of the product are you using? On what operating system?
pyminuit 1.2.0 linked against SEAL-Minuit 1.7.9 with python 2.7 on Mac OS X 10.7
Would it be possible to extend test_minuit.py to also run and check hesse and
minos?
And to make that easily accessible as minuit.test()?
Original issue reported on code.google.com by Deil.Chr...@googlemail.com on 29 Aug 2011 at 11:07
Hi,
Sorry, I don't see anything wrong with these results. To check them, I plotted
the function f (defined in test_minuit.py) at the minimum, plus-or-minus one
sigma:
>>> from cassius import * # http://code.google.com/p/cassius/
>>> view(Curve(lambda x: f(x, 0.00085), xmin=2.00-3.00, xmax=2.00+3.00))
>>> view(Curve(lambda y: f(2.00, y), xmin=0.00085-1., xmax=0.00085+1.))
In both direction, the minimum is at the center of the plot and the function
increases by about one unit at the edges. The non-parabolic shape of the y
curve is the reason for the difference between MIGRAD/HESSE and MINOS.
-- JIm
Original comment by jpivar...@gmail.com on 30 Aug 2011 at 3:34
Original issue reported on code.google.com by
Deil.Chr...@googlemail.com
on 29 Aug 2011 at 11:07Attachments:
The text was updated successfully, but these errors were encountered: