Running JSL in background without any windows popping up

[#1]

Hello,

New JSL user here. I have a nested for loop where I am using a bivariate platform to extract Rsq for each pair and creating a concatenated data table which will eventually hold all Rsq values for all pairs. I am able to get the desired output in the end. However, while the script is running, I get temp tables to pop up and I cannot seem to make them invisible or close (post contenation to another table).

Is there a way to flip a switch and turn off any windows to pop up? If not, anyone have a suggestion how to not have the screen keep popping tables while running? Output of the script while running is in the attachment. 

Script is below

For( i = 1, i <= 5, i++,
For (j= 1, j <= 5, j++,
biv = dt << Bivariate(
Y( ParmColumns[i] ),
X( ParmColumns[j] ),
Fit Line( {Line Color( {212, 73, 88} )} ),
Invisible

);
rbiv = biv << Report;
tb = rbiv[Table Box( 1 )];
mcdt = tb << Make Combined Data Table << Show Window( 1 );
if (counter == 0 ,
combineddt = mcdt
);
if (counter > 0 ,
combineddt << Concatenate(
mcdt,
"Append to first table"
);
);
counter ++;
rbiv << Close Window;

mcdt << Invisible;
combineddt << Invisible;
wait(.1);
);

);

// Select and delete all rows that are empty
combineddt << Select Where(:"Column 1" != "RSquare") << Delete Rows;
combineddt << Select Where(:X == :Y) << Delete Rows;

//save stacked data table
combineddt << Save (Output_File );


// Close data table and do not save JMP tables. CSV files are already saved.
Close All( Data Tables, NoSave );

 

[#2 · Accepted solution]

@AmanS ,

You should look up "invisible"  for report wndows and invisible vs private for data tables.

Names Default To Here( 1 );
flag=random Integer(0,1);
dt = Open( "$SAMPLE_DATA/Big Class.jmp" );
if(flag==1,
         obj = dt << Bivariate( Y( :Weight ), X( :Height ), Invisible ),
       //else
		obj = dt << Bivariate( Y( :Weight ), X( :Height ) )
		
 ); //end if

[#3]

Thanks @gzmorgan0 

Appreciate you taking the time to respond to this query

[#4 · Accepted solution]

You can learn much from @gzmorgan0's solution. It is provides a general framework. In this specific case, though, I will use JMP interactively to get what I want. No explicit iteration is required as JMP almost calculates the desired result for you. Just one extra step at the end.

Actually, I wrote a script to do it but the comments should provide the path to the result that you want.

Names Default to Here( 1 );

// open example with several continuous numeric variables
dt = Open( "$SAMPLE_DATA/Fitness.jmp" ) ;

// obtain names of such variables
cols = dt << Get Column Names( Continuous );

// convert the names to references
cols = dt << Get Column Reference( cols );

// use the references to launch platform with list of correlations
mult = dt << Multivariate( Y( Eval( cols ) ), Pairwise Correlations );

// obtain new data table with correlations
dt = Report( mult )["Pairwise Correlations"][TableBox(1)] << Make Into Data Table;

// convert correlations into linear coefficient of determination
dt << New Column( "R Square", Numeric, Continuous,
	Values( (dt:Correlation << Get As Matrix)^2 )
);

[#5]

Thanks @Mark_Bailey . I like how streamlined the code is using the multivariate feature instead of what I was trying to do with a complicated nested for loop. Much to learn!

[#6]

Removing the wait(...) may be sufficient. That gives the OS time to show the window. Wait is usually not needed, or there is an alternative that doesn't need it, like <<runformulas. Sometimes you do need the window to show/update though.

edit:

instead of

mcdt << Invisible; // does not work this way

which @gzmorgan0  showed the right way to use, you probably want to close the tables as soon as they are no longer needed. Change it to 

close(mcdt,"nosave");

I'd still leave out the wait(...) .