I have a 1 by 1000 (1 row by 1000 columns) matrix that contain only 0 and 1 as their elements. How can I find how many times 1 is repeated 3 times consecutively.
If there are more than 3 ones then it is necessary to reset the counting. So 4 would be 3+1 and it counts as only one instance of 3 consecutive 1s but 6 would be 3+3 so it counts as two instances of having 3 consecutive 1s.
This approach finds the differences between when A goes from 0 to 1 (rising edge) and from 1 to 0 (falling edge). This gives the lengths of consecutive 1s in each block. Then divide these numbers by 3 and round down to get the number of runs of 3.
Padding A with a 0 at the start and end just ensures we have a rising edge at the start if A starts with a 1, and we have a falling edge at the end if A ends with a 1.
A = round(rand(1,1000));
% padding with a 0 at the start and end will make this simpler
B = [0,A,0];
rising_edges = ~B(1:end-1) & B(2:end);
falling_edges = B(1:end-1) & ~B(2:end);
lengths_of_ones = find(falling_edges) - find(rising_edges);
N = sum(floor(lengths_of_ones / 3));
Or in a much less readable 2 lines:
A = round(rand(1,1000));
B = [0,A,0];
N = sum(floor((find(B(1:end-1) & ~B(2:end)) - find(~B(1:end-1) & B(2:end))) / 3));
You can define your custom functions like below
v = randi([0,1],1,1000);
% get runs in cell array
function C = runs(v)
C{1} = v(1);
for k = 2:length(v)
if v(k) == C{end}(end)
C{end} = [C{end},v(k)];
else
C{end+1} = v(k);
end
end
end
% count times of 3 consecutive 1s
function y = count(x)
if all(x)
y = floor(length(x)/3);
else
y = 0;
end
end
sum(cellfun(#count,runs(v)))
Here is another vectorized way:
% input
n = 3;
a = [1 1 1 1 0 0 1 1 1 0 0 0 1 1 1 1 1 0 1 1 1 1 1 1 1]
% x x x x x = 5
% output
a0 = [a 0];
b = cumsum( a0 ) % cumsum
c = diff( [0 b( ~( diff(a0) + 1 ) ) ] ) % number of ones within group
countsOf3 = sum( floor( c/n ) ) % groups of 3
You like it messy? Here is a one-liner:
countsOf3 = sum(floor(diff([0 getfield(cumsum([a 0]),{~(diff([a 0])+1)})])/n))
I would like to construct a function
[B, ind] = extract_ones(A)
which removes some sub-arrays from a binary array A in arbitrary dimensions, such that the remaining array B is the largest possible array with only 1's, and I also would like to record in ind that where each of the 1's in B comes from.
Example 1
Assume A is a 2-D array as shown
A =
1 1 0 0 0 1
1 1 1 0 1 1
0 0 0 1 0 1
1 1 0 1 0 1
1 1 0 1 0 1
1 1 1 1 1 1
After removing A(3,:) and A(:,3:5), we have the output B
B =
1 1 1
1 1 1
1 1 1
1 1 1
1 1 1
which is the largest array with only ones by removing rows and columns of A.
As the fifteen 1's of B corresponds to
A(1,1) A(1,2) A(1,6)
A(2,1) A(2,2) A(2,6)
A(4,1) A(4,2) A(4,6)
A(5,1) A(5,2) A(5,6)
A(6,1) A(6,2) A(6,6)
respectively, or equivalently
A(1) A(7) A(31)
A(2) A(8) A(32)
A(4) A(10) A(34)
A(5) A(11) A(35)
A(6) A(12) A(36)
so, the output ind looks like (of course ind's shape does not matter):
ind = [1 2 4 5 6 7 8 10 11 12 31 32 34 35 36]
Example 2
If the input A is constructed by
A = ones(6,3,4,3);
A(2,2,2,2) = 0;
A(4,1,3,3) = 0;
A(1,1,4,2) = 0;
A(1,1,4,1) = 0;
Then, by deleting the minimum cuboids containing A(2,2,2,2), A(4,1,3,3), A(1,1,4,3) and A(1,1,4,1), i.e. after deleting these entries
A(2,:,:,:)
A(:,1,:,:)
Then the remaining array B will be composed by 1's only. And the ones in B corresponds to
A([1,3:6],2:3,1:4,1:3)
So, the output ind lists the subscripts transformed into indices, i.e.
ind = [7 9 10 11 12 13 15 16 17 18 25 27 28 29 30 31 33 34 35 36 43 45 46 47 48 49 51 52 53 54 61 63 64 65 66 67 69 70 71 72 79 81 82 83 84 85 87 88 89 90 97 99 100 101 102 103 105 106 107 108 115 117 118 119 120 121 123 124 125 126 133 135 136 137 138 139 141 142 143 144 151 153 154 155 156 157 159 160 161 162 169 171 172 173 174 175 177 178 179 180 187 189 190 191 192 193 195 196 197 198 205 207 208 209 210 211 213 214 215 216]
As the array needed to be processed as above is in 8-D, and it should be processed more than once, so can anyone give me opinions on how to composing the program doing this task fast?
My work so far [Added at 2 am (GMT-4), 2nd Aug 2017]
My idea was that I delete the sub-arrays with the largest proportion of zero one by one. And here is my work so far:
Inds = reshape(1:numel(A),size(A)); % Keep track on which 1's survive.
cont = true;
while cont
sz = size(A);
zero_percentage = 0;
Test_location = [];
% This nested for loops are for determining which sub-array of A has the
% maximum proportion of zeros.
for J = 1 : ndims(A)
for K = 1 : sz(J)
% Location is in the form of (_,_,_,...,_)
% where the J-th blank is K, the other blanks are colons.
Location = strcat('(',repmat(':,',1,(J-1)),int2str(K),repmat(',:',1,(ndims(A)-J)),')');
Test_array = eval(strcat('A',Location,';'));
N = numel(Test_array);
while numel(Test_array) ~= 1
Test_array = sum(Test_array);
end
test_zero_percentage = 1 - (Test_array/N);
if test_zero_percentage > zero_percentage
zero_percentage = test_zero_percentage;
Test_location = Location;
end
end
end
% Delete the array with maximum proportion of zeros
eval(strcat('A',Test_location,'= [];'))
eval(strcat('Inds',Test_location,'= [];'))
% Determine if there are still zeros in A. If there are, continue the while loop.
cont = A;
while numel(cont) ~= 1
cont = prod(cont);
end
cont = ~logical(cont);
end
But I encountered two problems:
1) It may be not efficient to check all arrays in all sub-dimensions one-by-one.
2) The result does not contain the most number of rectangular ones. for example, I tested my work using a 2-dimensional binary array A
A =
0 0 0 1 1 0
0 1 1 0 1 1
1 0 1 1 1 1
1 0 0 1 1 1
0 1 1 0 1 1
0 1 0 0 1 1
1 0 0 0 1 1
1 0 0 0 0 0
It should return me the result as
B =
1 1
1 1
1 1
1 1
1 1
1 1
Inds =
34 42
35 43
36 44
37 45
38 46
39 47
But, instead, the code returned me this:
B =
1 1 1
1 1 1
1 1 1
Inds =
10 34 42
13 37 45
14 38 46
*My work so far 2 [Added at 12noon (GMT-4), 2nd Aug 2017]
Here is my current amendment. This may not provide the best result.
This may give a fairly OK approximation to the problem, and this does not give empty Inds. But I am still hoping that there is a better solution.
function [B, Inds] = Finding_ones(A)
Inds = reshape(1:numel(A),size(A)); % Keep track on which 1's survive.
sz0 = size(A);
cont = true;
while cont
sz = size(A);
zero_percentage = 0;
Test_location = [];
% This nested for loops are for determining which sub-array of A has the
% maximum proportion of zeros.
for J = 1 : ndims(A)
for K = 1 : sz(J)
% Location is in the form of (_,_,_,...,_)
% where the J-th blank is K, the other blanks are colons.
Location = strcat('(',repmat(':,',1,(J-1)),int2str(K),repmat(',:',1,(ndims(A)-J)),')');
Test_array = eval(strcat('A',Location,';'));
N = numel(Test_array);
Test_array = sum(Test_array(:));
test_zero_percentage = 1 - (Test_array/N);
if test_zero_percentage > zero_percentage
eval(strcat('Testfornumel = numel(A',Location,');'))
if Testfornumel < numel(A) % Preventing the A from being empty
zero_percentage = test_zero_percentage;
Test_location = Location;
end
end
end
end
% Delete the array with maximum proportion of zeros
eval(strcat('A',Test_location,'= [];'))
eval(strcat('Inds',Test_location,'= [];'))
% Determine if there are still zeros in A. If there are, continue the while loop.
cont = A;
while numel(cont) ~= 1
cont = prod(cont);
end
cont = ~logical(cont);
end
B = A;
% command = 'i1, i2, ... ,in'
% here, n is the number of dimansion of A.
command = 'i1';
for J = 2 : length(sz0)
command = strcat(command,',i',int2str(J));
end
Inds = reshape(Inds,numel(Inds),1); %#ok<NASGU>
eval(strcat('[',command,'] = ind2sub(sz0,Inds);'))
% Reform Inds into a 2-D matrix, which each column indicate the location of
% the 1 originated from A.
Inds = squeeze(eval(strcat('[',command,']')));
Inds = reshape(Inds',length(sz0),numel(Inds)/length(sz0));
end
It seems a difficult problem to solve, since the order of deletion can change a lot in the final result. If in your first example you start with deleting all the columns that contain a 0, you don't end up with the desired result.
The code below removes the row or column with the most zeros and keeps going until it's only ones. It keeps track of the rows and columns that are deleted to find the indexes of the remaining ones.
function [B,ind] = extract_ones( A )
if ~islogical(A),A=(A==1);end
if ~any(A(:)),B=[];ind=[];return,end
B=A;cdel=[];rdel=[];
while ~all(B(:))
[I,J] = ind2sub(size(B),find(B==0));
ih=histcounts(I,[0.5:1:size(B,1)+0.5]); %zero's in rows
jh=histcounts(J,[0.5:1:size(B,2)+0.5]); %zero's in columns
if max(ih)>max(jh)
idxr=find(ih==max(ih),1,'first');
B(idxr,:)=[];
%store deletion
rdel(end+1)=idxr+sum(rdel<=idxr);
elseif max(ih)==max(jh)
idxr=find(ih==max(ih),1,'first');
idxc=find(jh==max(jh),1,'first');
B(idxr,:)=[];
B(:,idxc)=[];
%store deletions
rdel(end+1)=idxr+sum(rdel<=idxr);
cdel(end+1)=idxc+sum(cdel<=idxc);
else
idxc=find(jh==max(jh),1,'first');
B(:,idxc)=[];
%store deletions
cdel(end+1)=idxc+sum(cdel<=idxc);
end
end
A(rdel,:)=0;
A(:,cdel)=0;
ind=find(A);
Second try: Start with a seed point and try to grow the matrix in all dimensions. The result is the start and finish point in the matrix.
function [ res ] = seed_grow( A )
if ~islogical(A),A=(A==1);end
if ~any(A(:)),res={};end
go = true;
dims=size(A);
ind = cell([1 length(dims)]); %cell to store find results
seeds=A;maxmat=0;
while go %main loop to remove all posible seeds
[ind{:}]=find(seeds,1,'first');
S = [ind{:}]; %the seed
St = [ind{:}]; %the end of the seed
go2=true;
val_dims=1:length(dims);
while go2 %loop to grow each dimension
D=1;
while D<=length(val_dims) %add one to each dimension
St(val_dims(D))=St(val_dims(D))+1;
I={};
for ct = 1:length(S),I{ct}=S(ct):St(ct);end %generate indices
if St(val_dims(D))>dims(val_dims(D))
res=false;%outside matrix
else
res=A(I{:});
end
if ~all(res(:)) %invalid addition to dimension
St(val_dims(D))=St(val_dims(D))-1; %undo
val_dims(D)=[]; D=D-1; %do not try again
if isempty(val_dims),go2=false;end %end of growth
end
D=D+1;
end
end
%evaluate the result
mat = prod((St+1)-S); %size of matrix
if mat>maxmat
res={S,St};
maxmat=mat;
end
%tried to expand, now remove seed option
for ct = 1:length(S),I{ct}=S(ct):St(ct);end %generate indices
seeds(I{:})=0;
if ~any(seeds),go=0;end
end
end
I tested it using your matrix:
A = [0 0 0 1 1 0
0 1 1 0 1 1
1 0 1 1 1 1
1 0 0 1 1 1
0 1 1 0 1 1
0 1 0 0 1 1
1 0 0 0 1 1
1 0 0 0 0 0];
[ res ] = seed_grow( A );
for ct = 1:length(res),I{ct}=res{1}(ct):res{2}(ct);end %generate indices
B=A(I{:});
idx = reshape(1:numel(A),size(A));
idx = idx(I{:});
And got the desired result:
B =
1 1
1 1
1 1
1 1
1 1
1 1
idx =
34 42
35 43
36 44
37 45
38 46
39 47
Say I have an array the size 100x150x30, a geographical grid 100x150 with 30 values for each grid point, and want to find consecutive elements along the third dimension with a congruous length of minimum 3.
I would like to find the maximum length of consecutive elements blocks, as well as the number of occurrences.
I have tried this on a simple vector:
var=[20 21 50 70 90 91 92 93];
a=diff(var);
q = diff([0 a 0] == 1);
v = find(q == -1) - find(q == 1);
v = v+1;
v2 = v(v>3);
v3 = max(v2); % maximum length: 4
z = numel(v2); % number: 1
Now I'd like to apply this to the 3rd dimension of my array.
With A being my 100x150x30 array, I've come this far:
aa = diff(A, 1, 3);
b1 = diff((aa == 1),1,3);
b2 = zeros(100,150,1);
qq = cat(3,b2,b1,b2);
But I'm stuck on the next step, which would be: find(qq == -1) - find(qq == 1);. I can't make it work.
Is there a way to put it in a loop, or do I have to find the consecutive values another way?
Thanks for any help!
A = randi(25,100,150,30); %// generate random array
tmpsize = size(A); %// get its size
B = diff(A,1,3); %// difference
v3 = zeros(tmpsize([1 2])); %//initialise
z = zeros(tmpsize([1 2]));
for ii = 1:100 %// double loop over all entries
for jj = 1:150
q = diff([0 squeeze(B(ii,jj,:)).' 0] == 1);%'//
v = find(q == -1) - find(q == 1);
v=v+1;
v2=v(v>3);
try %// if v2 is empty, set to nan
v3(ii,jj)=max(v2);
catch
v3(ii,jj)=nan;
end
z(ii,jj)=numel(v2);
end
end
The above seems to work. It just doubly loops over both dimensions you want to get the difference over.
The part where I think you were stuck was using squeeze to get the vector to put in your variable q.
The try/catch is there solely to prevent empty consecutive arrays in v2 throwing an error in the assignment to v3, since that would remove its entry. Now it simply sets it to nan, though you can switch that to 0 of course.
Here's one vectorized approach -
%// Parameters
[m,n,r] = size(var);
max_occ_thresh = 2 %// Threshold for consecutive occurrences
% Get indices of start and stop of consecutive number islands
df = diff(var,[],3)==1;
A = reshape(df,[],size(df,3));
dfA = diff([zeros(size(A,1),1) A zeros(size(A,1),1)],[],2).'; %//'
[R1,C1] = find(dfA==1);
[R2,C2] = find(dfA==-1);
%// Get interval lengths
interval_lens = R2 - R1+1;
%// Get max consecutive occurrences across dim-3
max_len = zeros(m,n);
maxIDs = accumarray(C1,interval_lens,[],#max);
max_len(1:numel(maxIDs)) = maxIDs
%// Get number of consecutive occurrences that are a bove max_occ_thresh
num_occ = zeros(m,n);
counts = accumarray(C1,interval_lens>max_occ_thresh);
num_occ(1:numel(counts)) = counts
Sample run -
var(:,:,1) =
2 3 1 4 1
1 4 1 5 2
var(:,:,2) =
2 2 3 1 2
1 3 5 1 4
var(:,:,3) =
5 2 4 1 2
1 5 1 5 1
var(:,:,4) =
3 5 5 1 5
5 1 3 4 3
var(:,:,5) =
5 5 4 4 4
3 4 5 2 2
var(:,:,6) =
3 4 4 5 3
2 5 4 2 2
max_occ_thresh =
2
max_len =
0 0 3 2 2
0 2 0 0 0
num_occ =
0 0 1 0 0
0 0 0 0 0
What is the best way of creating a 10x2 matrix in matlab where each element is a random int between 1-5, and so that there are only unique pairs of elements in this array? I know randperm can give a me random unique numbers, but Im not sure if its possible to use randperm to give unique pairs? The only other way I can think of is using:
randi([1 5], 10, 2);
In a loop with an if statement checking whether all pairs are unique.
An example of the data I would like would be something like:
4 5
1 3
2 2
1 4
3 3
5 1
5 5
2 1
3 1
4 3
Note: the order of the elements does not matter for example, both 4, 5 and 5, 4 would be valid.
First generate all possible pairs as rows of a matrix, then use randperm to generate a random subset of row indices:
N = 5; %// alphabet size
M = 2; %// number of columns
P = 10; %// desired number of rows
allPairs = dec2base(0:N^M-1, N)-'0'+1; %// generate all possible rows
ind = randperm(size(allPairs,1)); %// indices for random permutation of rows
ind = ind(1:P); %// pick P unique indices
result = allPairs(ind,:); %// use those indices to select rows
Example result:
result =
3 2
1 4
3 5
4 1
1 3
1 2
2 4
3 4
5 5
1 5
Here's another approach using randperm & dec2base without the memory overhead of generating all possible rows (quoting Luis's solution) -
%// Inputs
start = 1
stop = 5
Nr = 10 %// Number of rows needed
Nc = 2 %// Number of cols needed
intv = stop - start + 1; %// Interval/range of numbers
rand_ID = randperm(power(intv,Nc)-1,Nr); %// Unique IDs
out = dec2base(rand_ID,intv) - '0'+ start %// 2D array of unique numbers
Sample runs -
Case #1 (Same parameters as listed in question) :
start =
1
stop =
5
Nr =
10
Nc =
2
out =
1 3
2 1
5 3
5 4
5 5
3 4
2 3
2 5
3 3
1 4
Case #2 (Different parameters) :
start =
1025
stop =
1033
Nr =
10
Nc =
5
out =
1030 1029 1033 1028 1029
1033 1029 1026 1025 1025
1028 1026 1031 1028 1030
1028 1031 1027 1028 1025
1033 1032 1031 1029 1032
1033 1029 1030 1027 1028
1031 1025 1032 1027 1025
1033 1033 1025 1028 1029
1031 1033 1025 1033 1029
1028 1025 1027 1028 1032
Based on excaza's comment I found this to suit my needs:
n = randperm(5);
k = 2;
data = nchoosek(n, k);
Which gives some example output:
2 3
2 4
2 1
2 5
3 4
3 1
3 5
4 1
4 5
1 5