#1222 value() statement in fit .. via value() is not evaluated

[Karl Ratzsch]

If the variable name in a fit command is given as "value('varname'), statement is not evaluated, and an error message

Singular matrix in Givens()
error during fit

appears, when the same variable name without the "value" statement works. more a feature than a bug, no doubt, but i guess it´s an easy fix anyway.
gp is official 4.6.0 windows build

example:

a=2

fit a "-" via value("a")
1 2
1 3
1 2
e

[Ethan Merritt]

Your fit command makes no sense.
"fit a" is already incorrect, because a is not a function,
and the "via" part is also incorrect because it wants the
name of a variable, not a numeric value.

[Karl Ratzsch]

Am 15.03.2013 16:31, schrieb Ethan Merritt:

Your fit command makes no sense.
"fit a" is already incorrect, because a is not a function,

"fit a" works, and should work. You can define the function to be fitted
in the fit command itself, and f(x) = a is a valid function.

and the "via" part is also incorrect because it wants the
name of a variable, not a numeric value.

You´re right. I was too proud of having found a way emulate array
variables to notice that. ;-)

Then it it probably not a two-liner to get that to work. File it as a
feature request?

Regards, Karl

---

[bugs:#1222] http://sourceforge.net/p/gnuplot/bugs/1222/ value()
statement in fit .. via value() is not evaluated

Status: open
Created: Fri Mar 15, 2013 01:27 PM UTC by kf2402
Last Updated: Fri Mar 15, 2013 01:27 PM UTC
Owner: nobody

If the variable name in a fit command is given as "value('varname'),
statement is not evaluated, and an error message

Singular matrix in Givens()
error during fit

appears, when the same variable name without the "value" statement
works. more a feature than a bug, no doubt, but i guess it´s an easy fix
anyway.
gp is official 4.6.0 windows build

example:

a=2

fit a "-" via value("a")
1 2
1 3
1 2
e

---

Sent from sourceforge.net because you indicated interest in
https://sourceforge.net/p/gnuplot/bugs/1222/

To unsubscribe from further messages, please visit
https://sourceforge.net/auth/prefs/

--
Karl-Friedrich Ratzsch (Dipl. Chem.)
Freiburger Materialforschungszentrum / Universität Freiburg
Stefan-Meier-Straße 21, 79104 Freiburg im Breisgau
Tel. 0761/203-4748 Fax:-4701
ratzsch@uni-freiburg.de