sum separate continuous values

Hello Matlab Senior Users

I have a column of data 0 and 1 and the example data is like this format. There are 2 number of continous 1 value group. I don't want all groups of 1 values. I need only the number of continuous 1 value groups. It is 2 for this sample data (index 3 to 4 and index 6 to 8).

0
0
1
1
0
1
1
1
0
1
0
1
0
0

I tried some code like this suggested by some expert. But the answer is not correct.

ids=find(x==1);
choice=diff(ids)~=1;
ids([true;choice])
indexes=ids([true;choice]);
groups=numel(indexes)

If someone kindly advice me in for loop or while loop if will helpful for me.

Thank you very much for your helps.

Answer 1

[per isakson]

%%
d   = diff( c );
ixb = find( d == +1 ) + 1;
ixe = find( d == -1 );
%%
is_group = ixe >= ixb + 1;
[ ixb(is_group), ixe(is_group) ]

outputs

ans =
     3     4
     6     8
>> 

where

• c is the column of your question
• the first column of ans are indicies of the first "one" in a group and the second column the last.

Test this script with other input columns.

[soe thiha]

Hello per isakson

Thanks for your kind advice.

I want to get is the total number of continuous/consecutive groups of 1 values. There are only 2 groups of continuous 1 values (index 3 to 4 and index 6 to 8). I don't want to get all 1 values groups.

[per isakson]

I don't understand your comment.

"There are only 2 groups of continuous 1 values (index 3 to 4 and index 6 to 8)." Isn't that exactly what my script returns?

[soe thiha]

Hello Sir

Yes your codes is exactly perfet. And, I am sorry for my misunderstanding.

But, if I want to get the total number of groups, how can I do that,Sir.

Thank you for your kind helps.

Regards,

[per isakson]

"total number of groups" that's given by the number of rows of the output matrix. Thus, use

number_of_groups = size( output_of_script, 1 );

[soe thiha]

Dear Per Isakson

Thank you very much for your helps.

If you don't mind, may I request a next small question please?

Now I got the number of continuous 1 values groups. But, if I want to get the maximum value corresponding to the continuous groups, how can I do that please? I cannot use the value that is corresponding to 0 or individual 1 values. I have to find the max value within the continuous 1 groups .Follow is the sample data in which the answer is 2234.18.

  value  index 
  849.85  0 
  799.95  1 
  946.62  1 
  801.16  1 
  857.09  1 
  829.44  0 
  1569.19  0 
  2234.79  1 
  2168.86  0 
  2234.18  1 
  1556.63  1 
  962.04  0 
  838.04  1 
  774.14  1 
  827.39  1 
  754.04  1 
  857.09  1 
  915.24  1 

[per isakson]

%%
d   = diff( value_index(:,2) );
ixb = find( d == +1 ) + 1;
ixe = find( d == -1 );
%%
if ixb(end) > ixe(end)  % the last group includes the last row of value_index
    ixe(end+1) = size( value_index, 1 );
end
%%
is_in_group = false( size(value_index,1), 1 );
for jj = 1 : numel( ixb )
    is_in_group( ixb(jj) : ixe(jj) ) = true;
end
%%
max_value_in_groups = max( value_index( is_in_group, 1 ) );

where value_index is the table of your comment.