How to replace zero with the values present in array

Hello Everyone, I hope you are doing well. I have an array named as Dataset. I have replace 0 in for those values which has histogram count less than 50 percent of maximum value.

Now in my new array Dataset Has some Values which are equal to 0; I want to replace 0 with the Values already present in the array.

For example the array consists of value

34.9609375000000
37.9882812500000
34.9609375000000

These Values should be in placed of 0's

How can i do it in Matlab

Batchdata=Dataset;
fig=figure; set(fig,'visible','off');
h=histogram(Batchdata,10000);

sumofbins=max(h.Values);

size_MP=round(50/100*sumofbins);
ValueofHistogram= h.Values;
Bindata=h.Data;
Binedges=h.BinEdges;
Binedges(end) = Inf;
deleted_data_idx = false(size(Bindata));
for i=1: length(ValueofHistogram)
    if ValueofHistogram(i)<size_MP;
        deleted_data_idx(Bindata >= Binedges(i) & Bindata < Binedges(i+1)) = true;
    end
end
close(fig);


Dataset(deleted_data_idx,:)=0;

[Matt J]

For example the array consists of value ... These Values should be in placed of 0's

Your example isn't actually an example. It doesn't show us a hypothetical starting Dataset nor the final transformed one.

[Stephen john]

@Matt J It is actually example. 0 should be replaced by the the values which are present in the dataset . How can i do that?

[Matt J]

So, if

array=[34.9609375000000 37.9882812500000 34.9609375000000];

 DataSet=[100 800 0 900];

then the result should be

 newDataSet=[100 800 34.9609375000000 37.9882812500000 34.9609375000000  900];

I replace each zero with the entirety of array?

[Stephen john]

@Matt J No, each zero Replace with single value

For examples

DataSet=[100 800 0 900 0 100 0 0];

Then New Dataset is

DataSet=[100 800 800 900 900 100 800 900];

[Stephen john]

@Matt J There is second method. We will delete, the 0's values and repeat the array from the start to complete. same size as original

For examples I have 8 entries in the dataset

DataSet=[100 800 0 900 0 100 0 0];

New dataset after deleted 0; we have four entries

DataSet=[100 800 900 100 ];

and at last repeat the matrix to complete 8 entries

DataSet=[100 800 900 100 100 800 900 100];

Answer 1

[dpb]

array=[34.9609375000000
    37.9882812500000
    34.9609375000000];
Dataset(deleted_data_idx,:)=repmat(array,1,size(Dataset,2));

presuming

numel(array) == sum(deleted_data_idx)

[Stephen john]

@dpb The array values array=[34.9609375000000

37.9882812500000

34.9609375000000]; are already in the dataset.

[Stephen john]

@dpb There is second method. We will delete, the 0's values and repeat the array from the start to complete. same size as original

For examples I have 8 entries in the dataset

DataSet=[100 800 0 900 0 100 0 0];

New dataset after deleted 0; we have four entries

DataSet=[100 800 900 100 ];

and at last repeat the matrix to complete 8 entries

DataSet=[100 800 900 100 100 800 900 100];

[dpb]

DataSet=[100 800 0 900 0 100 0 0];
N1=numel(DataSet);
DataSet=DataSet(find(DataSet));
N2=numel(DataSet);
>> DataSet=[DataSet DataSet(1:N1-N2)]
DataSet =
   100   800   900   100   100   800   900   100
>> 

Above works if N2 >= N1/2

If N2 < N1/2, will have to reproduce remainder DataSet ceil(N1/N2) times and then keep N1 terms. Will leave exact code as "exercise for student"...

[dpb]

I have no klew what the first was intended to be, then...don't understand the specific problem trying to solve; too convoluted an explanation to try to delve into.

[Stephen john]

@dpb How to make this code generalized?

Because the first apporach is not working. So i changed the apporach, Thats why i asked

[dpb]

Which code is that? The last algorithm is general with the caveat noted that you've got to duplicate more than once to recreate the original length if you remove more than half the original elements.

[Stephen john]

@dpb How can ceil. Can we do it using repmat command?

Above works if N2 >= N1/2

If N2 < N1/2, will have to reproduce remainder DataSet ceil(N1/N2) times and then keep N1 terms. Will leave exact code as "exercise for student"...

[dpb]

"...do it using repmat command?"

How many times (but only as many times as needed)????

[Stephen john]

@dpb repeat to complete 1000 samples

[dpb]

You're missing the point -- how to calculate that number of repetitions is where the hint and ceil() come into play.