Istvan Dekany

I am building a composite predictive model that uses UMAP as an embedding component. I implemented the model in a class that stores trained UMAP objects as instance attributes. When...

Semi-supervised dimensionality reduction with continuous-valued labels (regression problem)

1

Hi, From the documentation, I understand how UMAP supports semi-supervised problems if the `target_metric` is `categorical` (default), i.e., by masking unlabelled data with -1. I have used this feature with...

What is the best way to regularize supervised UMAP?

1

I am working on a regression problem where I am attempting to use UMAP for supervised feature embedding, and use the resulting low-dimensional embeddings as input variables of a subsequent...