error code on cumsum

7 views (last 30 days)
Ralph
Ralph on 23 Feb 2014
Commented: Ralph on 23 Feb 2014
I am attempting to run: A = 1 - cumsum(C)/sum(C); where C is a long column vector. I receive error "??? Error using ==> cumsum CUMSUM is not supported for integer input." What does it mean?
  2 Comments
per isakson
per isakson on 23 Feb 2014
Edited: per isakson on 23 Feb 2014
The code looks ok. Make this check
which cumsum -all
what does it say? And what does
class( C )
say? And which Matlab release do you use?
Ralph
Ralph on 23 Feb 2014
response to 'which cumsum -all' EDU>> which cumsum -all built-in (C:\Program Files (x86)\MATLAB\R2010a Student\toolbox\matlab\datafun\@single\cumsum) % single method
built-in (C:\Program Files (x86)\MATLAB\R2010a Student\toolbox\matlab\datafun\@double\cumsum) % double method
built-in (C:\Program Files (x86)\MATLAB\R2010a Student\toolbox\matlab\datafun\@logical\cumsum) % logical method
built-in (C:\Program Files (x86)\MATLAB\R2010a Student\toolbox\matlab\datafun\@uint8\cumsum) % uint8 method
built-in (C:\Program Files (x86)\MATLAB\R2010a Student\toolbox\matlab\datafun\@int8\cumsum) % int8 method
built-in (C:\Program Files (x86)\MATLAB\R2010a Student\toolbox\matlab\datafun\@uint16\cumsum) % uint16 method
built-in (C:\Program Files (x86)\MATLAB\R2010a Student\toolbox\matlab\datafun\@int16\cumsum) % int16 method
built-in (C:\Program Files (x86)\MATLAB\R2010a Student\toolbox\matlab\datafun\@uint32\cumsum) % uint32 method
built-in (C:\Program Files (x86)\MATLAB\R2010a Student\toolbox\matlab\datafun\@int32\cumsum) % int32 method
Response to class (C) EDU>> class(C) ans = uint8
I have Matlab R2010a

Sign in to comment.

Accepted Answer

Image Analyst
Image Analyst on 23 Feb 2014
Edited: Image Analyst on 23 Feb 2014
Try this. It works for me:
C = int32(randi(9, 1, 10)) % Array of integers.
A = 1 - cumsum(C) / sum(C)
  3 Comments
Image Analyst
Image Analyst on 23 Feb 2014
Yes, MATLAB has an unintuitive of promoting/casting variable types. If C is double, it's fine. Regardless I can't find a way to duplicate the error the Ralph posted. Perhaps he can give some code that illustrates the error.
Ralph
Ralph on 23 Feb 2014
The suggestion above to use int32() provided an idea on how to proceed. i did not know how to identify what format my data was in, so the class() command helped with this. My data was uint8. Do I used double() to change the format of my data, and I got passed the cumsum error. Thanks for the assistance.

Sign in to comment.

More Answers (0)

Tags

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!