How to creat a script that generat a Pareto Plot with date filters

[#1]

Hello everyone,

I hope you're doing well. I'm facing an issue creating a script to generate a Pareto plot from a table that I want to filter by dates (date1 and date2).

As a first step, I called my table using:

dt2 = Open( "/D:/Project JMP/Tables/Table NOK PY.jmp", invisible );

Next :

dt2_top5 = dt2 << Summary(
	Group( :Date, :Ref Produit, :ProductName, :Baie, :ICT_FCT, :CodeError),
	Freq( "Aucun(e)" ),
	Weight( "Aucun(e)" ),
	output table name( "Table2 Top 5" ),
	invisible
);

// Subset of data for Pareto plot filtered by Date
dt_pareto_subset = dt2_top5 << Select Where( Num( Char( :Date ) ) >= Num( date1 ) & Num( Char( :Date ) ) < Num( date2 ) );
dt_pareto_subset = dt2_top5 << Subset(
	Selected Rows(1),
	Columns(:CodeError),
	output table name("Pareto Subset"),
	invisible
);

and i did creat the Pareto plot in my window  :

Border Box(Left(25), Top(50), Bottom(200),
	V List Box(
		// Add the Pareto plot here
		paretoPlot = dt_pareto_subset << Pareto Plot(
			Y( :CodeError ),
			Freq( "Freq" ),
			Show Percents( 1 ),
			Show Cumulative Line( 1 )
		),
	)	
)

but after i lanched my script i did get this error :

The script object is expected when accessing or evaluating "Send", dt_pareto_subset << /*###*/
Pareto Plot(
               Y(:CodeError),

               Freq("Freq"),

               Show Percents(1),

               Show Cumulative Line(1)

)/*###*/

I have a small code snippet that generates a Pareto plot. Currently, I have to manually select the data table each time, and I cannot apply a date filter : 

Pareto Plot(
	Cause( :CodeError ),
	SendToReport(
		Dispatch( {"Plots"}, "Pareto Report", FrameBox, {Frame Size( 1555, 725 )} )
	)
);

Thank you all for your help,

Best regards,
Yass

[#2]

Do you have the subset you expect in dt_pareto_subset?

[#3]

yes and i expect dt_pareto_subset to contain only the rows from dt2_top5 where the value in the :Date column falls between date1 and date2

[#4]

The error message is telling that the variable dt_pareto_subset doesn't have a scriptable object (in this case data table). There could many different causes for this and it is totally guess work without seeing your data or the full script.

[#5]

Names Default To Here( 1 );

TOP5 = New Namespace(
    "TOP5"
);

TOP5:get_dates = Function({},
    ////
);

TOP5:calculate_nok_percentages = Function({dt},
    /////
);

// Define the Do function
TOP5:Do = Function ({nb_date1, nb_date2},
    //Ouverture de la table de données
    dt = Open( "/D:/Project JMP/Tables/TableData PY.jmp", invisible );
    dt2 = Open( "/D:/Project JMP/Tables/Table NOK PY.jmp", invisible );


    {percent_nok_ict, percent_nok_fct} = TOP5:calculate_nok_percentages(dt);

    current_month = Format(Today(), "Format Pattern", "<Month>", << Use Locale(0));

    date1 = Date Increment(Today(), "day", nb_date1, "start");
    date2 = Date Increment(Today(), "day", nb_date2, "start");

    dt_top5 = dt << Summary(
        Group( :Date, :Nom de fichier, :Ref Produit, :ProductName, :Baie, :ICT_FCT, :Result, :CodeError),
        Freq( "Aucun(e)" ),
        Weight( "Aucun(e)" ),
        output table name( "Table Top 5" ),
        invisible
    );

    dt2_top5 = dt2 << Summary(
		Group( :Date, :Ref Produit, :ProductName, :Baie, :ICT_FCT, :CodeError),
		Freq( "Aucun(e)" ),
		Weight( "Aucun(e)" ),
		output table name( "Table2 Top 5" ),
		invisible
	);

	// Subset of data for Pareto plot filtered by Date
	dt_pareto_subset = dt2_top5 << Select Where( Num( Char( :Date ) ) >= Num( date1 ) & Num( Char( :Date ) ) < Num( date2 ) );
	dt_pareto_subset = dt2_top5 << Subset(
		Selected Rows(1),
		Columns(:CodeError),
		output table name("Pareto Subset"),
		invisible
	);


    /*
    //
    //
    //

    Here i have the logic of production repport.
    it's fine it works 
    
    //
    //
    //
    */


    //Création de la fenetre du top 5
    window = new Window ("TOP5",
        V List Box(
            pb = Picture Box( Open( "D:\Project JMP\Images\logo.png", png ) ),
            Border Box( Top(50),
                V List Box(
                    tb1 = Text Box("Production results report of "||Format( date2, "yyyy-mm-dd" )),
                    tb1 << Set Base Font( "Title" ),
                    tb1 << Set Font Scale (3),
                    tb1 << Justify Text("Center"),
                    tb1 << Set Width(2770),
                    tb1 << Background Color(RGB Color(225, 225, 225)),
                    
                    tb2 = Text Box ("SVI, PY"),
                    tb2 << Justify Text("Center") << Set Font Size(25),
                    tb2 << Set Width(2770),
                    tb2 << Background Color(RGB Color(240, 240, 240)),
                )
            ),
            V List Box (
                Border Box (Top (100), Bottom(100), Left (25),
                    H List Box(
                        // TOP 5 ICT
                        V List Box(
                            tb3 = Text Box( "Top 5 NOK results for ICT"),
                            tb3 << Set Base Font( "Title" ),
                            tb3 << Set Font Scale (2),
                            H List Box(
                                // column bay
                                
                                // column product ref
                                
                                // column test number
                                
                                //column %nok
                                
                                // column product name
                            )
                        ),	
                        V List Box(
							//nok percent for this month
						),
                    )
                ),
                Border Box(Left(25), Top(50), Bottom(200),
                    //TOP 5 FCT
                    V List Box(
                        tb4 = Text Box( "Top 5 NOK results for FCT"),
                        tb4 << Set Base Font( "Title" ),
                        tb4 << Set Font Scale (2),
                        H List Box(
                                // column bay
                                
                                // column product ref
                                
                                // column test number
                                
                                //column %nok
                                
                                // column product name
                            )
                        )
                        V List Box(
							//nok percent for this month
						),
                    )
                ),
                
                Border Box(Left(25), Top(50), Bottom(200),
					V List Box(
						// Add the Pareto plot here
						paretoPlot = dt_pareto_subset << Pareto Plot(
							Y( :CodeError ),
							Freq( "Freq" ),
							Show Percents( 1 ),
							Show Cumulative Line( 1 )
						),
					)	
				)
            )
        )
    );

    window << Maximize Window (1);
    //Enregistrement du fichier image .png et du fichier texte .txt
    window << Save Picture ("D:\Project JMP\Rapports de Production Quotidiens\Rapport du "||Format( date2, "yyyy-mm-dd" )||".png", "png");
    window << Save Text ("D:\Project JMP\Rapports de Production Quotidiens\Rapport du "||Format( date2, "yyyy-mm-dd" )||".txt");
);

[#6]

Does it work as intended if you create the plot after you have created the subset? So don't put it inside the new window

Names Default To Here(1);
dt = Open("$SAMPLE_DATA/Time Series/Steel Shipments.jmp");

dt << New Column("Errors", Character, Nominal, Formula(
	Match(Floor(:Steel Shipments / 1000),
		4, "1000",
		5, "2000",
		6, "3000",
		7, "4000",
		8, "5000"
	)
));

rows_of_interest = dt << Get Rows Where(02Feb1985<= :Date <= 09Jan1990);

dt_subset = dt << Subset(Rows(rows_of_interest), Selected Columns(0), Output Table("Subset"));

paretoplot = dt_subset << Pareto Plot(Cause(:Errors));

[#7]

Hello,

i tried the method in your script, it gives me the pareto plot but without data filter. For example the pareto for today must be 2024-07-29 <= :Date < 2024-07-30.

So I used this functions :

// Function to get today's and yesterday's date
GetTodayAndYesterday = Function({},
    today = Today();
    yesterday = Date Increment(today, "day", -1, "start");
    evallist({Format(yesterday, "yyyy-mm-dd"), Format(today, "yyyy-mm-dd")});
);

	dt = Open( "/D:/Project JMP/Tables/DB.jmp", invisible );

    {yesterday, today} = GetTodayAndYesterday();

     dt_top5 = dt << Summary(
		     Group( :Date, :Ref Produit, :Baie, :ICT_FCT, :Result, :CodeError),
		     Freq( "Aucun(e)" ),
		     Weight( "Aucun(e)" ),
		     output table name( "Table Top 5" ),
		     invisible
	     );

    // Create a subset of data for the Pareto Plot filtered by the date range
    rows_of_interest = dt_top5 << Get Rows Where( Num( Char( :Date ) ) >= Num( yesterday ) & Num( Char( :Date ) ) <= Num( today ) );
    dt_pareto_subset = dt_top5 << Subset(Rows(rows_of_interest), Selected Columns(0), Output Table("Pareto Subset"), invisible);

    // Generate the Pareto Plot
    paretoplot = dt_pareto_subset << Pareto Plot(Cause(:CodeError));	

I've attached a sample of my database for reference.

[#8]

Do you want to have a data filter in your pareto plot or do you just want to filter the data? It seems like you just want to create a subset from your data and NOT have a filter.

[#9]

I need to filter the data with Date using a subset with conditions i need, so i can create a Pareto plot for yesterday's production results because i generate a daily production resalts report every morning.

[#10 · Accepted solution]

There should be no need for the num and char to num conversions.

Maybe something like this?

Names Default To Here(1);

dt = Open("$DOWNLOADS/DB.jmp");

yday = Date Increment(Today(), "day", -1, "start");
tday = Date Increment(Today(), "day", 0, "start");


dt_top5 = dt << Summary(
	Group(:Date, :Ref Produit, :Baie, :ICT_FCT, :Result, :CodeError),
	Freq("Aucun(e)"),
	Weight("Aucun(e)"),
	output table name("Table Top 5")
);

rows_of_interest = dt_top5 << Get Rows Where(yday <= :Date <= tday);
show(rows_of_interest);

dt_pareto_subset = dt_top5 << Subset(
	Rows(rows_of_interest),
	Selected Columns(0),
	Output Table("Pareto Subset")
);

pp = dt_pareto_subset << Pareto Plot(Cause(:CodeError), Freq(:N Rows));

You could also create paretoplot from your initial table (filtering can require a bit of work)

dt << Pareto Plot(
	Cause(:CodeError),
	Local Data Filter(
		Add Filter(
			columns(:Date),
			Where(:Date == 3805056000),
			Display(:Date, N Items(4))
		)
	)
);