finding a numeric pattern in an array

Hi all,

I have a big numeric vector and I am trying to find a pattern in the vector.

Example:

my numeric vector = [11 5 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 11 2 31 5 1 2 11 2 31 5 1 2 11 2 31 5 1 2 11 2 31 5 1 2 11 2 31 5 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 11 2 31 5 1 2 11 2 31 5 1 2 11 2]

pattern : [1 2 3 4 5 *any numbers* 1 2 3 4 5 1 2 11 2 31]

I know strfind works well if I know the exact sequence of pattern. But for the above case I am unable to find the solution.

Answer 1

[Stephen23]

V = [11,5,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,11,2,31,5,1,2,11,2,31,5,1,2,11,2,31,5,1,2,11,2,31,5,1,2,11,2,31,5,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,3,4,1,2,3,4,5,1,2,3,4,5,1,2,3,4,5,1,2,11,2,31,5,1,2,11,2,31,5,1,2,11,2]

V = 1×133

11 5 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

F = @(v) regexptranslate('escape',char(v));
P = sprintf('%s.*?%s',F([1,2,3,4,5]),F([1,2,3,4,5,1,2,11,2,31]))

P = '□□□□□.*?□□□□□□□□□□'

[X,Y,M] = regexp(char(V),P, 'start','end','match') % start & end indices, matched text.

X = 1×2

6 64

Y = 1×2

35 122

M = 1×2 cell array
    {'□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□'}    {'□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□'}

Checking the matched content:

double(M{1})

ans = 1×30

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 11 2 31

double(M{2})

ans = 1×59

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

[Saikrishna]

@Stephen23 can't we just take char([1 2 3 4 5 '.*?' 1 2 3 4 5 1 2 11 2 31])?

[Stephen23]

"can't we just take char([1 2 3 4 5 '.*?' 1 2 3 4 5 1 2 11 2 31])?"

That should work the same as my original answer, but ... I realized that we also need to account for special characters that need escaping, so you might also need to use REGEXPTRANSLATE something like this:

F = @(v) regexptranslate('escape',char(v));
P = sprintf('%s.*?%s',F([1,2,3,4,5]),F([1,2,3,4,5,1,2,11,2,31]))

P = '□□□□□.*?□□□□□□□□□□'

I will update my answer with this.

[Saikrishna]

@Stephen23 excellent! Thank you :) that worked for me perfectly

Answer 2

[Dyuman Joshi]

vec=[11     5     3     4     5     1     2     3     4     5     1     2     3     4     5     1     2    3     4     5     1     2     3     4     5     1     2     3     4     5     1     2    11     2   31     5     1     2    11     2    31     5     1     2    11     2    31     5     1     2    11  2    31     5     1     2    11     2    31     5     3     4     5     1     2     3     4     5 1     2     3     4     5     1     2     3     4     5     1     2     3     4     5     1     2  3     4     5     1     2     3     4     5     1     2     3     4     5     1     2     3     4    1     2    3     4     5     1     2     3     4     5     1     2     3     4     5     1     2    11     2   31     5     1     2    11     2    31     5     1     2    11     2];
pat1 = 1:5;
pat2 = [pat1 1 2 11 2 31];
%Starting index of pattern 1
idx1=strfind(vec,string(pat1))

idx1 = 1×15

6 11 16 21 26 64 69 74 79 84 89 94 103 108 113

%Ending index of pattern 2
idx2=strfind(vec,string(pat2))+numel(pat2)-1

idx2 = 1×2

35 122

Any viable combination of elements of idx1 to idx2 will be the pattern i.e. 6-35, 11-35, 16-35, 6-122, 64-122, ...

A viable combination is idx1(m)<idx2(n)