topic Re: graph builder, local data filter, and large data sets in Discussions

graph builder, local data filter, and large data sets

bartimaeusx — Sat, 13 Dec 2025 00:22:06 GMT

Hello,

I have a dashboard of sorts using local data filter and graph builder to browse through a stacked dataset of measurement data (one VALUE column and a few measurement identifier columns. The functionality is great, but if my dataset is large, JMP freezes/crashes as it tries to jam all the data into the initial, unfiltered graph when the graph builder script is launched or the local data filter is change to "none selected" in the first conditional data filter.

So I am wondering if there is a way to suppress the evaluation of the graph if there is "too much" data is included in the graph, until a Order|Operation item is selected, or some other way to disallow a user from selecting some combination of data that results in jmp crashing as it tries to plot the entire dataset.

Thank you!

JMP pro 17.2

Re: graph builder, local data filter, and large data sets

jthi — Sat, 13 Dec 2025 07:06:06 GMT

I don't think you can prevent user from selecting "None Selected" but you can pre-build the graph builder in such a way that it will be at least fast on the first load but this does require some scripting. Maybe create a JMP Wish List item about this. Something along the lines of: when a single selection type display is used (Single Category / Radio Box) allow us disabling None Selected option. Or maybe even wider definition which would force user to always have something selected from the filters even with "multi-select" options?

If I really want to start optimizing for the speed of (local) data filter, I either: write my own using different display boxes OR I link the data filter to smaller table which is then used to filter the "main" table using << Make Filter Change Handler. Scripting is needed for both of these options.