Using App Designer: Pulling data from an X,Y indices from a user input in 2 drop down menus

Hello!

see my update below with errors

Okay, now I am getting these errors in the command window (may have been getting them the whole time, just didn't realize it). I'm off to see if I can figure this out, but any help would be greatly appreciated! Thanks!

Incorrect number or types of inputs or outputs for function 'FluxDataPulled'.

Error in matlab.apps.AppBase>@(source,event)executeCallback(ams,app,callback,requiresEventData,event) (line 62)

newCallback = @(source, event)executeCallback(ams, ...

Error while evaluating DropDown PrivateDropDownOpeningFcn.

Incorrect number or types of inputs or outputs for function 'FluxDataPulled'.

Error in matlab.apps.AppBase>@(source,event)executeCallback(ams,app,callback,requiresEventData,event) (line 62)

newCallback = @(source, event)executeCallback(ams, ...

Error while evaluating Button PrivateButtonPushedFcn.

Incorrect number or types of inputs or outputs for function 'FluxDataPulled'.

Error in matlab.apps.AppBase>@(source,event)executeCallback(ams,app,callback,requiresEventData,event) (line 62)

newCallback = @(source, event)executeCallback(ams, ...

Error while evaluating Button PrivateButtonPushedFcn.

end update

Attached is my .m file and my spreadsheet. I have my whole tool working, except I am trying to pull data from a spreadsheet, I have in the right file path. I do not get any errors when I run it, however, when I select the "Flux" button, nothing happens. The rest of the tool works as long as I manually enter in the flux. Below is a c/p of the code from this portion, and in the .m file it's in lines 39 through 87. I have a feeling i'm not "reading" the table correctly. I've tried many variations of this including:

Table = readtable('FluxDataInterpolated.xlsx','Sheet1',false);
Table = readtable('FluxDataInterpolated.xlsx','Sheet1',true);
Table = readtable('FluxDataInterpolated.xlsx','Sheet1','Data');

    properties (Access = private)
       
        inclination
        altitude
        flux
    end
    % Callbacks that handle component events
    methods (Access = private)
        % Code that executes after component creation
        function startupFcn(app)
            Table = readtable('FluxDataInterpolated.xlsx','Sheet1','Data');
            app.inclination = Table{1,2:end};
            app.altitude = Table{2:end,2};
            app.flux = Table{:,3}; 
            app.DisplayFlux.Value = app.flux(1);
        end
        % Drop down values for Inclination and Altitude
        function InclinationDegDropDownValueChanged(app, ~)
            xvalue = app.InclinationDegDropDown.Value;
            yvalue = app.AltitudekmDropDown.Value;
            xindices = find(app.inclination==str2double(xvalue));
            yindices = find(app.altitude==str2double(yvalue));
            idx=intersect(xindices,yindices);
            app.DisplayFlux.Value = app.flux(idx);
            
        end
        function AltitudekmDropDownValueChanged(app, ~)
            xvalue = app.InclinationDegDropDown.Value;
            yvalue = app.AltitudekmDropDown.Value;
            xindices = find(app.inclination==str2double(xvalue));
            yindices = find(app.altitude==str2double(yvalue));
            idx=intersect(xindices,yindices);
            app.DisplayFlux.Value = app.flux(idx);
            
        end
        % Push Flux button to Display Flux from table
        function Fluxdatapulled(app, ~)
            
            xvalue = app.InclinationDegDropDown.Value;
            yvalue = app.AltitudekmDropDown.Value;
            xindices = find(app.inclination==str2double(xvalue));
            yindices = find(app.altitude==str2double(yvalue));
            idx=intersect(xindices,yindices);
            app.DisplayFlux.Value = app.flux(idx);
        end
    end

Answer 1

[Walter Roberson]

Table = readtable('FluxDataInterpolated.xlsx', 'Sheet', 'Sheet1');

Note that the file only has one sheet, and the default for readtable is to read the one sheet, so you could just use

Table = readtable('FluxDataInterpolated.xlsx');

It is not clear to me why you are not using readmatrix as I had suggested before.

[Veronica Vigil]

Hi Walter - i did try readmatrix below, just forgot to list it as one of the things. I both Table = readtable('FluxDataInterpolated.xlsx'); and Table = readmatrix('FluxDataInterpolated.xlsx'); just now, and same results.

now I'm thinking it's not my read table/matrix function, it's something with the coding below and the push button. I'm not getting any errors at all, when I select the flux button it just doesn't do anything.

[Walter Roberson]

Did you know that you can configure ItemsData for drop down lists, and put numeric values there? So you might configure the items as {'100', '200', '300'...} but instead of using str2double() on that, configure ItemsData as [100, 200, 300, ...] and then the Value property will be that numeric value directly.

            xindices = find(app.inclination==str2double(xvalue));
            yindices = find(app.altitude==str2double(yvalue));
            idx=intersect(xindices,yindices);

Are you expecting vectors for xindices and yindices? I do not understand that call to intersect()

            app.DisplayFlux.Value = app.flux(idx);

You know that flux is a 2D array there, not a vector?

[Veronica Vigil]

They definitely aren't vectors. It's pulling the right data, because when I hover over this function (find(inclination==xvalue);, it's pulling the right inclinations from the right row. However, it's not pulling the data from the array. I'm just trying to pull the value that corressponds with an inclination (in 1,2,3...170) and Altitude (300, 301...1200), that value is a flux value and is in the form of 3.33 e-7 (ish).

The app.DisplayFlux.Value is the name of the box I designated to display the Flux value pulled. The app.flux(idx); is my attempt to display that flux. Worked in other parts of my code, even tried app.DisplayFlux.Value = idx; because that worked for me in another area, but no dice. Now it seems like the code is getting hung up somewhere and I'm getting "value is not a double scalar". attached is the code from app designer and is an mlapp format. when I run it, the GUI pops up, I change the drop down values for inclination and altitude, and the window goes behind all the others and an "abstractnumericcomponent.m" opens in main matlab window and the following message is displayed in the command window:

Error using Probofcoll/FluxDataPulled

'Value' must be a double scalar.

Error in matlab.apps.AppBase>@(source,event)executeCallback(ams,app,callback,requiresEventData,event) (line 62)

newCallback = @(source, event)executeCallback(ams, ...

188 throw(exceptionObject);

In workspace belonging to Probofcoll/FluxDataPulled (line 68)

[Veronica Vigil]

This is now the code I have. As you mentioned, it appears to be doing something funky with the intersect command. when I run the code in app designed, it gets hung up at the app.DisplayFlux.Value = Int; (line 14 below) with a green arrow next to this line in the command window of app designer. When I hover over the "xvalue" or "yvalue" on line 12 while the code is hung up, it is pulling the correct data from the userdrop downs, so that's functioning. When I hover over "xvar" and "yvar" in line 13 below, it says for xvar: "xvar: 1x1 double = 40" --- I have no idea what the value of 40 is. For "yvar" it says "yvar: 0.1 empty double column vector". I think this relates back to what @Walter Roberson said above where he asked me "you know 2D flux is an array and not a vector". I know that from the spreadsheet, but I'm guessing intersect(x,x) is not the right command and converts it to a vector? How do I pull just a plain old value if I have an X and a Y in a table? In theory, I want the "interection", but that's not working here. My matlab research is telling me this is the right function, but apparently it's not.

NOTE: Some of the beginning code from line 3 up was generated by the App Designer and is greyed out, so I cannot modify it.

1 % Callback function: AltitudekmDropDown, DisplayFlux, FluxButton,

2 % ...and 1 other component

3 function FluxDataPulled(app, event)

4 Table = readtable('FluxDataInterpolated.xlsx');

5 %inclination = app.InclinationDegDropDown.Value;

6 %altitude = app.AltitudekmDropDown.Value;

7 inclination = Table{1,2:end};

8 altitude = Table{2:end,1};

9 xvalue = app.InclinationDegDropDown.Value;

10 yvalue = app.AltitudekmDropDown.Value;

11 xvar = find(inclination==xvalue);

12 yvar = find(altitude==yvalue);

13 Int = intersect(xvar,yvar);

14 app.DisplayFlux.Value = Int;

end

[Walter Roberson]

Suppose that the user choose the third inclination value, and the fifth altitutude values. Then provided that you got lucky and the comparisons for floating point equality worked out (a risky assumption!) the first find() would return the scalar 3, and the second find() would return the scalar 5. You then ask to intersect(3, 5) and of course the result is empty most of the time. You then index flux() at that empty intersection.

Sometimes you would get lucky and the user chose the 3rd inclination and also the 3rd altitude. Then both find() would return 3, and intersect(3,3) would be 3... and then you would ask to pull out flux(3) . But your flux is a 2D array, so you would be pulling out flux(3,1) instead of flux(3,3)

You should not be doing the intersect(): you should be using the results of the find() as the direct indexes into flux()

[Veronica Vigil]

@Walter Roberson thank you again for your help. I tried this after reading your last response, I agree I don't want jsut an estimate. I want it to pull the value from the table of inclination 100 and altitude 500 = which in the table is 8.81384506886443E-07. I then want to display that in my app after you hit the flux button. I changed the code to the below. Which is obvioiusly wrong because it still does not work. I've been researching a good part of the day, and cannot figure this out.

xvar = find(inclination==xvalue);

yvar = find(altitude==yvalue);

Int = find(xvar,yvar);

app.DisplayFlux.Value = Int;

[claudio]

have you check the limits of your display flux?

[Walter Roberson]

            xvar = find(inclination==xvalue);

You expect that to return a scalar.

           yvar = find(altitude==yvalue);

You expect that to return a scalar.

           Int = find(xvar,yvar);

The first input to find() is expected to be a scalar there.

The basic operation of find() is to look for non-zero entries in the first parameter and return the indices of those non-zero values. find() never returns 0 because 0 is not a possible index.

Since xvar is the result of a find() call, xvar would be a list of zero or more positive values -- expected to be exactly one non-zero value. So none of the entries in xvar can be 0, and so when you ask to find the indices of the non-zero entries in xvar then that is going to be the indices of all (typically 1) entries in xvar.

When you pass a second parameter to find(), that limits the number of indices returned to be no more than the given number. For example, find(XYZ, 3) would ask for at most the first three non-zero entries of XYZ.

We are expecting xvar to be scalar, so we are expecting there to be only one entry to find the index of, so we are not expecting the limit being expresed through yvar to have any effect.

Taking again the example I did before: you might find() that the inclination of the drop-down box might correspond to row 3 of the table, and you might find() that the altitude corresponded to column 5 -- so you would be asking for Int = find(3, 5) which asks for at most the first 5 indices of the non-zero entries in the scalar 3 . We would therefore expect that Int will store 1 almost all of the time.

What can change that? Well, if inclination and xvalue are character vectors then you would likely have trouble. Consider that '1' == '10' is not going to return a scalar false and would instead return the vector [true false] -- because == between character vectors tries to compare each character against the respective character in the other vector if both entries are non-scalar, and if one of the entries is scalar then that scalar will be compared against each entry in the other. '10' is not a scalar, it is a vector, ['1' '0']

Do not use == to compare entries that are intended to represent groups. Use strcmp() or strcmpi() . Or make sure you are using string() objects: "1" == "10" does check to see whether the group 1 matches the group 10 as a whole.

But none of this would be necessary if you were to do what I recommended earlier and configured the ItemsData values to be directly the row number or column number to be used: then you would use the values directly as indices.

[Veronica Vigil]

@Walter Roberson Thank you for all your help! I was able to figure it out.