CJ Carey

You could define a `TransposedCOO` wrapper class that re-writes axes before passing through to the underlying COO class. This is essentially the approach we use for CSR/CSC matrices in scipy.sparse.

Also cc @pv, @ev-br. I agree that forcing sparse methods to return the same format is ultimately going to be a performance hit, while the workarounds for users who want...

I suppose you could make a format similar to CSR called CDR (compressed-dense-row) where you'd only have `indptr` and `data` arrays, and each nonzero row would be fully dense. I'm...

FWIW, this is the approach I used for sparray: https://github.com/perimosocordiae/sparray/blob/master/sparray/flat.py#L20 I found it to be quite convenient, though it does mean you may need more integer bits to represent the...

Yes, I've wanted to fix this issue with the test suite since gh-51. It'll be difficult to get the right test coverage without over-specifying results, but any progress on this...

I'm not exactly sure what you're asking, but if you'd like to open a Pull Request with your changes I'll be happy to take a look.

The first issue is that the k-stochastic-neighbor repo doesn't have a clear license, so you'll probably want to ask @kswersky for permission before copying that code here. (For reference, this...

For an example of conditionally building native code, see: https://github.com/perimosocordiae/sparray/blob/e13fb0948b0f3b38813f072031d97dcf2dcd57e3/setup.py#L4

Can you explain a little more on the inputs and outputs of the `TupleTransformer`? It seems like it would need access to the label information at some point, but I'm...

I'm not sure if there's a standard approach in the literature for this sort of thing, but one thing you could try right now is using your existing distance to...