It has no transform option

[jochenater]

For reproducibility, it should also have a transform option, so it can transform datapoint which is hasn't been trained on.
On top of that, to be able to use the sklaern Pipeline, it needs this functionality as well

[eamid]

Currently, there is no clear way of extending the embedding to new datapoints. For new points, you need to run TriMap from scratch. I'm thinking about different ways of adding an option in the future to support new datapoints (i.e. approximately extending the embedding).