I have a heat map that needs some nested groups (in order to properly show values inside the heat map) squares. My issue is that a lot of the rows are entirely missing because it forces every group to exist. Is there a way to only show valid groups? For example:
Names Default to here(1);
dt = open("$SAMPLE_DATA\Big Class.jmp");
rows = dt << Get Rows Where(:age == 14 & :sex == "M");
dt << delete rows(rows);
dt << Graph Builder(
Variables( Group Y( :sex ), Group Y( :age ), Color( :height ) ),
Elements( Heatmap( Legend( 4 ) ), Caption Box( Legend( 5 ) ) )
);
Gives.
I don't want to see the 14 year old Male category.
Or, if you know a better way to show the means for a column inside the square, I'd take that too.
If you have JMP 14, heatmap now has a labeling capability. You may not need to use groups:
Would labeled bars work?
dt << Graph Builder( Size( 456, 413 ), Show Control Panel( 0 ), Variables( X( :height ), Y( :sex ), Y( :age, Position( 1 ) ), Color( :height ) ), Elements( Bar( X, Y( 1 ), Y( 2 ), Legend( 5 ), Label( "Label by Value" ) ) ), SendToReport( Dispatch( {}, "sex", ScaleBox, {Reversed Scale} ) ) );
Going to Bump this because it has come up again. Anyone have thoughts on a good way to do this?
If you have JMP 14, heatmap now has a labeling capability. You may not need to use groups:
Awesome! Now I just have to convince them to upgrade. :-P
More than appreciated John,
V
You may find some convincing material here: Explore These New Features in JMP® 14.