Subscribe Bookmark RSS Feed

local data filter problem?

ram_asra_gmail_

Community Trekker

Joined:

Mar 21, 2013

Hi everyone,

i am using local data filter to select items of legend to compare them. in this i have rows which are treated as targets (which i want to keep them as black color and dashed fit line). so i do that manually after fitting the lines. now when i start using local data filter, color of target data points changed to blue and all red and blue fit lines changed to black as well. as you can see in that in bottom picture.

i started using data filter because i want to hide fit line as well as data points when you select a legend since lines are not dynamic only data is.

i also have pasted code for data table which will have this chart too, if any one want to see the problem.

10570_before datafilter.PNG

after data filter

10571_after data filter.PNG

New Table( "Test",

  Add Rows( 16 ),

  New Script(

  "Bivariate",

  Bivariate(

  Y( :b ),

  X( :a ),

  Automatic Recalc( 1 ),

  Fit Where(

  :c == "A",

  Fit Orthogonal(

  Fit X to Y,

  {Line Color( "Medium Dark Red" ), Line Width( 2 )}

  )

  ),

  Fit Where(

  :c == "B",

  Fit Orthogonal(

  Fit X to Y,

  {Line Color( "Medium Dark Green" ), Line Width( 2 )}

  )

  ),

  Fit Where(

  :c == "tgt",

  Fit Orthogonal(

  Fit X to Y,

  {Line Color( "Black" ), Line Style( "Dashed" ), Line Width( 2 )}

  )

  ),

  SendToReport(

  Dispatch(

  {},

  "1",

  ScaleBox,

  {Min( 0 ), Max( 9.23175965665236 ), Inc( 2 ), Minor Ticks( 0 ),

  Label Row Nesting( 1 )}

  ),

  Dispatch( {}, "2", ScaleBox, {Max( 14.1371428571429 )} ),

  Dispatch(

  {},

  "Bivar Plot",

  FrameBox,

  {DispatchSeg( Line Seg( 1 ), {Line Width( 2 )} ),

  DispatchSeg( Line Seg( 2 ), {Line Width( 2 )} ),

  DispatchSeg(

  Line Seg( 3 ),

  {Line Style( "Dashed" ), Line Width( 2 )}

  ), Row Legend(

  c,

  Color( 1 ),

  Color Theme( "JMP Dark" ),

  Marker( 0 ),

  Marker Theme( "" ),

  Continuous Scale( 0 ),

  Reverse Scale( 0 ),

  Excluded Rows( 0 )

  )}

  )

  )

  )

  ),

  New Column( "a",

  Numeric,

  "Continuous",

  Format( "Best", 13 ),

  Set Values( [1, 2, 3, 4, 5, 6, 7, 8, 1, 2, 3, 4, 5, 6, 7, 8] )

  ),

  New Column( "b",

  Numeric,

  "Continuous",

  Format( "Best", 13 ),

  Set Values( [2, 3, 5, 6, 6, 7, 9, 10, 4, 5, 6, 7, 8, 9, 10, 11] )

  ),

  New Column( "c",

  Character,

  "Nominal",

  Set Values(

  {"A", "A", "B", "B", "A", "A", "B", "B", "tgt", "tgt", "tgt", "tgt",

  "tgt", "tgt", "tgt", "tgt"}

  )

  ),

  Set Row States(

  [4864, 4864, 5120, 5120, 4864, 4864, 5120, 5120, 5376, 5376, 5376, 5376,

  5376, 5376, 5376, 5376]

  )

)

2 REPLIES
Mandy_JMP

Staff

Joined:

Jan 14, 2015

Just want to confirm, what version of JMP are you using?

I think if you make a change to the section in your script related to Legend, then selecting a color change in the legend area works better.

      {Row Legend(

                          sex,

                          Color( 0 ),

                          Color Theme( "" ),

                          Marker( 0 ),

                          Marker Theme( "" ),

                          Continuous Scale( 0 ),

                          Reverse Scale( 0 ),

                          Excluded Rows( 0 )

                    )}

When I made these changes then I was able to change the "tgt" value to black and have it remain black.

ram_asra_gmail_

Community Trekker

Joined:

Mar 21, 2013

Hi , with color theme(""), how you will have color theme applied to legends? for a new table rowstates will always be empty and all data points are black for that.