Counting number of runs (excluding zeros)?

Hi there, I'm looking for a bit of help to count the number of runs of numbers but excluding zeros.

Scoured the internet but just cant find something that ignores zeros.

Bascially I have a column of data made up of 1's, -1's, and 0's

I need to count the number of runs of 1's and the number of runs of '-1's.

A = [0,1,1,1,1,1,0,0,0,1,1,1,0,1,1,1,0,-1,-1,-1,-1,-1,-1,0,0,1,1,-1,-1,0]

The data would like like this, and the answer would be 6!

Thanks for your help :)

[madhan ravi]

Illustration

[Aaron Varma]

A = [0,1,1,1,1,1,0,0,0,1,1,1,0,1,1,1,0,-1,-1,-1,-1,-1,-1,0,0,1,1,-1,-1,0]

The data would like like this, and the asnwer would be 6!

Thanks

Answer 1

[Alan Stevens]

Try plus1 = sum(data>0) and neg1 = sum(data<0)

[Aaron Varma]

Thanks Alan but i think this would just add up all the 1's whereas I need to know the number of seperate run's of 1's. I've updated my query to make that more clear.

Thanks

Answer 2

[Image Analyst]

You need bwlabel() to stamp each contiguous region with an ID number. Then add them together:

A = [0,1,1,1,1,1,0,0,0,1,1,1,0,1,1,1,0,-1,-1,-1,-1,-1,-1,0,0,1,1,-1,-1,0]
[~, num1s] = bwlabel(A == 1)
[~, numMinus1s] = bwlabel(A == -1)
totalRegions = num1s + numMinus1s

[Aaron Varma]

Thanks, I believe this is an add-on to the basic package, is that correct?

I haven't got that yet, do you know is there another way to do this task?

Thanks

[Image Analyst]

It's in the Image Processing Toolbox, which almost everyone has. The closest thing is findgroups() but I'm not sure how to make it count the groups. Maybe if you use splitapply() or something. If you need to write your own bwlabel() routine, I'm sure there are "connected components labeling" code out there.

[Aaron Varma]

Thanks for your help, i'll have a look into it, when using bwlabel i get a message about not having th right package so guess i dont have it yet.

Thanks