Skip to content

Add cluster export and t-SNE caching - #542

Open
FBartos wants to merge 1 commit into
masterfrom
codex/clustering-export-enhancements
Open

Add cluster export and t-SNE caching#542
FBartos wants to merge 1 commit into
masterfrom
codex/clustering-export-enhancements

Conversation

@FBartos

@FBartos FBartos commented Aug 8, 2026

Copy link
Copy Markdown

Adds export support for clustering results (predicted cluster, soft memberships, and t-SNE coordinates) with new helper functions (.mlClusteringExportResultsToData, AddCluster/SoftMemberships/TSNE helpers and caching).
Computes soft memberships for fuzzy-cmeans, model-based and RF (from proximity) and normalizes them. Moves t-SNE to reusable cached routine handling duplicate rows.
Fixes randomForest predict calls to pass only predictor columns (classification/regression/DALEX explain).

Adds export support for clustering results (predicted cluster, soft memberships, and t-SNE coordinates) with new helper functions (.mlClusteringExportResultsToData, AddCluster/SoftMemberships/TSNE helpers and caching). Computes soft memberships for fuzzy-cmeans, model-based and RF (from proximity) and normalizes them. Moves t-SNE to reusable cached routine handling duplicate rows. Fixes randomForest predict calls to pass only predictor columns (classification/regression/DALEX explain). Updates QML UI to expose export options and updates .gitignore/.Rbuildignore for .positai/.claude.
@FBartos
FBartos requested a review from koenderks August 8, 2026 12:57
@github-actions

github-actions Bot commented Aug 8, 2026

Copy link
Copy Markdown
Contributor

👋 Friendly reminder: It looks like NEWS.md wasn't updated in this Pull Request.

If your changes include bug fixes, new features, or UI tweaks, please consider adding a quick note to the # jaspYourModule (development version) section so our users know about your awesome work! (If this is just a minor typo fix, feel free to ignore this message.)

@koenderks

Copy link
Copy Markdown
Collaborator

Couple questions without looking at the code yet:

  1. We already export the predicted cluster, right?

  2. What is the use case for exporting the T-sne coordinates? I'm struggling to see one because they don't mean anything and are not fixed for a dataset. Weighing that with potential UI clutter makes me thing whether we really need it.

@FBartos

FBartos commented Aug 9, 2026

Copy link
Copy Markdown
Author
  1. The cluster prediction yes, not it also exports the probabilties.
  2. We got request for the t-SNE export because its apparently useful to have the values if you want to plot them outside of the module/use for later analyses (I do not remmember the exact usecase). (and since I was doing that, I thought that the soft membership might also be nice to have).

I agree with the clutter worry, thats why I changed the group into a collapsed section now:
image
In many other analyses we are often asked to export more values into the dataset (i.e., residuals, random effects etc... since people wanna do custom things with them) and having a separate section for those things is quite nice.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants