Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
Drexel University or The Math Forum.
|
|
|
|
Re: trouble saving data structure to a file
Posted:
Nov 6, 2012 10:20 AM
|
|
"Ben Ruppel" <brspam@cox.net> wrote in message
news:k7b6ao$q01$1@newscl01ah.mathworks.com...
> "Steven_Lord" <slord@mathworks.com> wrote in message
> <k7a66a$asf$1@newscl01ah.mathworks.com>...
*snip*
> Thank you very much, your solution worked. I'm not thrilled with creating
> a copy of A, B, and C since these are very large variables.
Depending on what you're doing, you may not be creating an actual copy.
http://www.mathworks.com/help/matlab/matlab_prog/memory-allocation.html
The copy-on-write behavior may save you from actually making a copy.
> How can you tell that handles.inputData is not a scalar struct variable?
Variable names in MATLAB have four specific rules:
http://www.mathworks.com/help/matlab/ref/isvarname.html
'handles.inputData' contains a period, which is not allowed in variable
names. That makes it an expression that when evaluated returns a variable.
If you assign the output of that expression to an identifier, that
identifier must be a valid variable name.
> In Variable Editor both handles and handles.inputData are listed as "1x1
> struct". Furthermore, they are both data "things" in matlab which contain
> multiple data structures and arrays and cell arrays. What makes the one
> different from the other?
"Ben Ruppel" is the name you use to post to CSSM.
"The person that posts to CSSM using the address 'brspam@cox.net' " refers
to you, but it is not your name. If I "evaluate" that expression by
searching CSSM, I can find your actual name. [Not the best of examples, as
anyone can spoof their email address, but hopefully good enough for
illustration purposes.]
> Is there a way for me to make handles.inputData a scalar struct variable?
For purposes of SAVE? Not without assigning it to a variable.
> Or is the problem more that handles.inputData is an expression? If that
> is the case, is there a way to point to this data structure that does not
> count as an expression?
If you're worried about memory, see above.
--
Steve Lord
slord@mathworks.com
To contact Technical Support use the Contact Us link on
http://www.mathworks.com
|
|
|
|
