How to round values of table with mixed types?

I would like to round all values in given columns, but get the error To assign to or create a variable in a table, the number of rows must match the height of the table.

I have attached a photo of the workspace variable of the table, only want to round certain rows as others have values which are unable to round

varnames={'Set Temp','duration','Start Temp','Time to Temp','Time to Steady','overshoot'};
for k=1:height(MasterTable)
    for m=1:length(varnames)
        varname=varnames{m};
        MasterTable{k,varname}=mat2cell(round(cell2mat(MasterTable{k,varname})),3);
    end
end

[dpb]

'Set Temp' appears to be integer-valued already excepting you've got some locations that contain an array instead of a single value. Why is that; that'll screw up trying to do anything with the variable. Fix the data design first, then you can simply write

t.(varname)=round(t.(varname));

and be done.

[Nicholas Kavouris]

Some cycles have more than one set temp, this is not a viable solution as this data design is necessary information to my programs outputs

Answer 1

[Steven Lord]

Using a sample table:

Temp = [37.3;39.1;42.3];
Pressure = [30.1;30.03;29.9];
WindSpeed = [13.4;6.5;7.3];
WindDirection = categorical({'NW';'N';'NW'});
T = table(Temp,Pressure,WindSpeed,WindDirection)

T = 3×4 table
    Temp    Pressure    WindSpeed    WindDirection
    ____    ________    _________    _____________

    37.3      30.1        13.4            NW      
    39.1     30.03         6.5            N       
    42.3      29.9         7.3            NW      

Let's round all the numeric variables.

T2 = varfun(@round, T, 'InputVariables', @isnumeric)

T2 = 3×3 table
    round_Temp    round_Pressure    round_WindSpeed
    __________    ______________    _______________

        37              30                13       
        39              30                 7       
        42              30                 7       

We can see which variables from T were used in the varfun computation.

areNumeric = varfun(@isnumeric, T, 'OutputFormat', 'uniform')

areNumeric = 1×4 logical array
   1   1   1   0

We could use those to overwrite the contents of T with T2. I'm going to make a backup copy of the original T for use in one more step.

T3 = T;
T(:, areNumeric) = T2

T = 3×4 table
    Temp    Pressure    WindSpeed    WindDirection
    ____    ________    _________    _____________

     37        30          13             NW      
     39        30           7             N       
     42        30           7             NW      

If you only want to round some of the values, specify InputVariables differently.

v = [true false true false];
T4 = varfun(@round, T3, 'InputVariables', v); % leave Pressure alone
T3(:, v) = T4

T3 = 3×4 table
    Temp    Pressure    WindSpeed    WindDirection
    ____    ________    _________    _____________

     37       30.1         13             NW      
     39      30.03          7             N       
     42       29.9          7             NW      

Answer 2

[Voss]

If those columns that contain numeric values are cell arrays, as in

FW = ["EVT64"; "EVT64"; "EVT64"];
SetTemp = {[400;225]; [203;194]; 450};
Duration = {104.95; 264.15; 29.9833};

MasterTable = table(FW,SetTemp,Duration)

MasterTable = 3×3 table
      FW         SetTemp         Duration  
    _______    ____________    ____________

    "EVT64"    {2×1 double}    {[104.9500]}
    "EVT64"    {2×1 double}    {[264.1500]}
    "EVT64"    {[     450]}    {[ 29.9833]}

then you can modify your code as follows:

varnames = {'SetTemp','Duration'};
for k=1:height(MasterTable)
    for m=1:length(varnames)
        varname=varnames{m};
%         MasterTable{k,varname}=mat2cell(round(cell2mat(MasterTable{k,varname})),3);
        MasterTable.(varname){k} = round(MasterTable.(varname){k});
    end
end
MasterTable

MasterTable = 3×3 table
      FW         SetTemp       Duration
    _______    ____________    ________

    "EVT64"    {2×1 double}    {[105]} 
    "EVT64"    {2×1 double}    {[264]} 
    "EVT64"    {[     450]}    {[ 30]}