AnsweredAssumed Answered

AW: vrf linear Regression problem: goodness 1

Question asked by rsb on Dec 8, 2005
"Baranski, Detlef <> wrote:
> I did some more trials.
> Setting each data to real 64 did not help.
> Have a look for the error showed in the pictures.
> I try to extract the code and the data.
> The seldom thing is that with the extracted code and the stored data it is
> running....
> So how can I store the data on a "line" in a way without changing it ?

pipe it into a data constant. That way it gets stored with the program
and can be used later.

> I think it was also a wish for the new Vee8 to be able to store datas
> directly when a program is buggy just by clicking the line and additionally
> to the "showing" of the data you should be able to store the data.

yep- that would be slick. I've also long requested the ability
to determine whether or not the data on a given line/terminal is valid.
So perhaps a valid/invalid switch along with a "save" button
on the line probe.

If you compress a VEE program it gets a whole bunch smaller, so
you might ZIP it using max compression settings. Then mail it
or if it is really big post it somewhere for downloads.

As for the errors- you have a divide by zero somewhere- hence the overflow.

>From what I can see of your error messages the problem is inside
of your "regression" function. If this is a user object we need to
see what's inside it.

Otherwise, what is the full value of the inputs to the regression

Also- it might make little sense to use a full regression routine
if you only have 2 data points. The regression/interpolation for
two points would be much simpler with a one-line formula ( think
2 pt form of a straight line ).



Stan Bischof  Agilent Technologies  707-577-3994

You are currently subscribed to vrf as:
To subscribe send a blank email to "".
To unsubscribe send a blank email to "".
To send messages to this mailing list,  email "". 
If you need help with the mailing list send a message to "".
Search the "unofficial vrf archive" at "".