MATLAB: Removing all-NaN Rows/Columns from Cell Matrix

Question

Let's say I have the following cell matrix that contains NaN's, numbers, and strings…

A = {...
  0    2    'All'  NaN  NaN  2
  NaN  NaN  NaN    NaN  NaN  NaN
  NaN  NaN  'Yes'  NaN  NaN  1};

Is there an easy way to identify the rows and columns that contain all NaN's? In this example, I would like to remove row 2, and then columns 4 and 5. The result would be…

B = {...
    0    2    'All'  2
    NaN  NaN  'Yes'  1};

I have brute-forced it by checking every value, but I'm sure there is a more methodical approach using cellfun perhaps. My method took 30 lines but I'm guessing there it can be done in just a few lines.

Thanks!

Answer 1

A = {  0    2    'All'  NaN  NaN  2
NaN  NaN  NaN    NaN  NaN  NaN
NaN  NaN  'Yes'  NaN  NaN  1}
idx=cellfun(@(x) all(isnan(x)),A)
ii=all(idx,2)
jj=all(idx)
A(ii,:)=[]
A(:,jj)=[]