필터 지우기
필터 지우기

Using strcmp on multiple strings to get a logical array

조회 수: 200 (최근 30일)
Marc Cousoulis
Marc Cousoulis 2017년 1월 12일
편집: James Tursa 2017년 1월 16일
I want to check for the occurrence of 3 possible events in a series of events..
Event = ['a1','b1','a1','c1','b1']';
Check = {'a1','b1','c1'}';
LogA(:,1) = strcmp(Event,Check{1});
LogA(:,2) = strcmp(Event,Check{2});
LogA(:,3) = strcmp(Event,Check{3});
This gives me a 5x3 logical array, but is there a way to do it without calling each column individually? i.e., what if I had 100 strings I wanted to check for and the goal was to have a 5x100 logical array? I don't want to use a "for-loop". I'm assuming there is some matrix way to do this.
  댓글 수: 1
Jan
Jan 2017년 1월 15일
Note:
Event = ['a1','b1','a1','c1','b1']';
is the same as:
Event = ('a1b1a1c1b1').';
Most likely you mean a cell string instead.

댓글을 달려면 로그인하십시오.

채택된 답변

Kirby Fears
Kirby Fears 2017년 1월 12일
편집: Kirby Fears 2017년 1월 12일
There's really nothing wrong with a for loop in this case. However, cellfun is probably what you're looking for:
Event = {'a1','b1','a1','c1','b1'}';
Check = {'a1','b1','c1'}';
l = cellfun(@(c)strcmp(c,Event),Check,'UniformOutput',false);
This produces a cell array the size of Check with logical arrays the size of Event.
You can index like this:
l{index1}(index2)
  댓글 수: 2
Stephen23
Stephen23 2017년 1월 13일
편집: Stephen23 2017년 1월 13일
Also simple to convert to a logical array:
>> Event = {'a1','b1','a1','c1','b1'};
>> Check = {'a1','b1','c1'};
>> X = cellfun(@(c)strcmp(c,Event),Check,'UniformOutput',false);
>> vertcat(X{:})
ans =
1 0 1 0 0
0 1 0 0 1
0 0 0 1 0
Jan
Jan 2017년 1월 15일
편집: James Tursa 2017년 1월 16일
strcmp does not accept: strcmp(Event, Check.') and this does not work also:
bsxfun(@strcmp, Event, Check.') % ERROR!
What a pity.

댓글을 달려면 로그인하십시오.

추가 답변 (2개)

Guillaume
Guillaume 2017년 1월 13일
편집: Guillaume 2017년 1월 13일
Another option which may or may not be faster than cellfun(...):
Event = {'a1','b1','a1','c1','b1'}';
Check = {'a1','b1','c1'}';
LogA = false(numel(Event), numel(Check));
[~, loc] = ismember(Event, Check);
LogA(sub2ind(size(LogA), 1:numel(Event), loc.')) = true

John BG
John BG 2017년 1월 13일
Marc
1.
If all 2nd halves of the strings contained in Event is '1', it is reasonable to ignore the entire column:
Event = ['a1';'b1';'a1';'c1';'b1'];
Event_num=double(Event)
=
97 49
98 49
97 49
99 49
98 49
E=Event_num(:,1)';
E =
97 98 97 99 98
2.
Same applies to Checks
Check = ['a1';'b1';'c1'];
Check_num=double(Check)
=
97 49
98 49
99 49
C=Check_num(:,1)';
C =
97 98 99
3.
Occurence of events are listed in variable occ
[Lie,occ]=ismember(E,C)
Lie =
1 1 1 1 1
occ =
1 2 1 3 2
4.
check
Check(occ)
=
abacb
meaning occ contains the occurrences: a1 b1 a1 c1 b1
if you find these lines useful would you please mark my answer as Accepted Answer?
To any other reader, if you find this answer of any help please click on the thumbs-up vote link,
thanks in advance for time and attention
John BG
  댓글 수: 5
Guillaume
Guillaume 2017년 1월 14일
As is clearly explained in the help page, answers can only be accepted by somebody else after 7 days of inactivity by the original author.
Jan
Jan 2017년 1월 15일
편집: Jan 2017년 1월 16일
@John BG: Marc can still decide, even if he knows why your solution in not efficient. Other readers and you can profit from this information also.

댓글을 달려면 로그인하십시오.

카테고리

Help CenterFile Exchange에서 Characters and Strings에 대해 자세히 알아보기

태그

제품

Community Treasure Hunt

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

Start Hunting!

Translated by