Incremental median across pages of a 3D array

조회 수: 2(최근 30일)
Matt J
Matt J 2021년 1월 25일
편집: Matt J 2021년 1월 28일
I am trying to compute the inter-page median,
of a 3D array A except that A is too large to be held in memory in its entirety and its pages A(:,:,k) occupy separate files on disk (EDIT: and I do not have the means to read in strict subchunks of a page). Is there an algorithm, and ideally also a Matlab implementation somewhere, that will compute B incrementally by looping over successive pages A(:,:,k), or chunks of pages?
  댓글 수: 4
Matt J
Matt J 2021년 1월 28일
That sounds like it would require representing each page as a datastore. I don't have the means to load in subsections of a page, e.g, individual rows, so I don't see how such a datastore could be set up.

댓글을 달려면 로그인하십시오.


Gaurav Garg
Gaurav Garg 2021년 1월 28일
Hi Matt,
You can compute the median of each column by converting the column into tall column and then calculating its median. You can repeat the step for each column (in your case).
answer = gather(m);
Or, you can also convert the array into distributed array and then compute median (though the former solution might be more useful).
D = distributed(A);
e = median(D);
  댓글 수: 1
Matt J
Matt J 2021년 1월 28일
Hi Gaurav,
The idea is to obtain the median of A(i,j,:) for each fixed pair (i,j). What you propose seems to require converting the 3D array into a 2D array, permuted and reshaped somehow so that the pages are now columns. But I don't see how you would accomplish that since, as I said, the pages, A(:,:,k) canot be held simultaneously in RAM, nor are they available contiguously on disk.

댓글을 달려면 로그인하십시오.

Matt J
Matt J 2021년 1월 28일
편집: Matt J 2021년 1월 28일
One solution would be to decimate and concatenate the pages, as below. If I choose a modest stride, it becomes possible to store the decimated pages simultaneously in RAM and take their median,
for i=1:stride
for j=1:stride
for k=1:numPages
temp=read(___); %read k-th page
B(i:stride:end,j:stride:end)=median( cat(3,Asubsets{:}) , 3);
However, this approach requires stride^2 passes through the files and a lot of discarded page data in each pass. So, I was hoping for a method that could be done in only a single pass, eliminating the outer two loops.


Community Treasure Hunt

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

Start Hunting!

Translated by