can't fill missings on table

Hi,

when running

opts = detectImportOptions(filename);
opts.SelectedVariableNames={'TimeStamp','RAW_AF7','RAW_AF8','RAW_TP9','RAW_TP10','AUX_RIGHT'};
T = readtable(filename,opts,'ReadVariableNames',true);
summary(T)
Variables:

    TimeStamp: 64x1 cell array of character vectors

    RAW_AF7: 64x1 double

        Values:

            Min        4.0293  
            Median     602.99  
            Max          1650  

    RAW_AF8: 64x1 double

        Values:

            Min            0.80586 
            Median         405.75  
            Max            1650    
            NumMissing     2       

    RAW_TP9: 64x1 double

        Values:

            Min            7.2527  
            Median         879.6   
            Max            1650    
            NumMissing     3       

    RAW_TP10: 64x1 double

        Values:

            Min             24.982  
            Median          865.7   
            Max             1650    
            NumMissing      4       

    AUX_RIGHT: 64x1 double

        Values:

            Min              0       
            Median           825.6   
            Max              1650    
            NumMissing       3       

t = fillmissing(T,'pchip');

I'm getting:

Error using fillmissing/checkArrayType (line 507)
Table contains variables of types for which the 'pchip'
fill method is not supported.

Error in fillmissing/fillTableVar (line 166)
[intConstVj,extMethodVj] =
checkArrayType(Avj,intMethod,intConstVj,extMethodVj,x,true);

Error in fillmissing/fillTable (line 144)
        B.(vj) =
        fillTableVar(indVj,A.(vj),intMethod,intConst,extMethod,x,useJthFillConstant,useJthExtrapConstant);
        
Error in fillmissing (line 127)
        B =
        fillTable(A,intM,intConstOrWinSize,extM,x,dataVars); 

what i'm doing wrong?

Here is an example file with 64 rows: csv file

I've tried converting the table to cell, and then every entry to numeric or string, and the back to table, but the error persist.

Answer 1

[Peter Perkins]

Apparently at least one of the table variables that you are applying fillmissing to isn't numeric. You will have to figure out which one.

That link doesn't seem very helpful. You should not need to be converting things back and forth. Provide a short eaxmple file, and a short example of what readtable is giving you back.

[kira]

I will update the answer with new info. All rows with missing values are double. I found the problem, I had to remove the TimeStamp column. How do I put it back?

[kira]

nevermind, i figure out! :)