James Melville

`ret_model` isn’t supported with a distance matrix because it’s primarily intended for transforming new data and you need the actual original input data to find the distances between them and...

Ok, that sounds like it would be possible but I can't say when (or if) it will get done.

@luciat-92 does #64 cover your use case?

In the sample code, the rows of `X` are basically identical, at least up to the 8 decimal places that my Jupyter Lab notebook displays, so the small variation in...

The [rcppannoy issues](https://github.com/eddelbuettel/rcppannoy/issues) would be a better place to ask for help about RcppAnnoy. As far as I remember, while multi-threading is available for building indices (but not searching) in...

Hello, thanks for the request. I'll look at inverse transforms when the UMAP Python package has an official release supporting it, so that I know the implementation is somewhat stable....

Hello @edwindj, thank you for the link to the paper, which I had not seen. Certainly it is something I will keep in mind, depending on how easy it is...

Thank you for the detailed response, @bwlewis. I would certainly be interested in testing an improved `partial_eigen` routine in uwot.

It looks like Seurat have embedded a version of the R wrapper script into their code (see [here](https://github.com/satijalab/seurat/blob/7618ac8798e2aefc509529f07af9d2836b96a1be/R/dimensional_reduction_internal.R#L327)). It doesn't seem to work with version 1.1.0 of FIt-SNE. You might...

I am very late to discovering this, but yes this is almost certainly because of the underlying UMAP package ([uwot](https://cran.r-project.org/package=uwot)), which uses [RcppAnnoy](https://cran.r-project.org/package=RcppAnnoy), which itself wraps the C++ library [Annoy](https://github.com/spotify/annoy)...