t-SNE implementation in Knime


Did anyone successfully created or attempted to use t-SNE in Knime?

Is t-SNE in Knime's future plans?


Sounds pretty niche-y to me, but interesting. R snippet nodes are your friend for such things. :-)


Hi amoldovan,

as E already pointed out, this would be a use case for the R or Python nodes.

Best, Iris

Indeed :) thanks!

Hi everyone,

I use t-SNE all the time in KNIME through the Rtsne package because the 2D/3D node scatterlplot node is awesome, but I think it is definitely mainstream enough that I would think about implementing it if I was on the KNIME team. In fact, a new subset of the Mining branch could be dimension reduction, and it could have t-SNE, diffusion maps, PCA etc. Just a thought.


I 100% agree with Brian.

There is a PCA node in Knime which I found it useful a few times.


I just discovered t-SNE references on the DL4J website, in the Word2Vec section no less:


So I suppose it's not quite that much of a niche usage anymore, given that Word2Vec is now (almost) part of "Standard KNIME"... :-)


We have such a node on our roadmap. This will be part of the DL4J extension then. However, I can't promise that it will be part of the upcoming release.

No rush, R will serve the purpose fine until then. :)

Let us know if there is any advances on this node.



Last reply was a year ago, and the comment from DaveK was such a node is on the roadmap. Any further on this? Would definitely be useful.



1 Like



Brilliant, thanks. A good reason for upgrading to 4. :smile:


1 Like

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.