marmaduke woodman

> Pro: libhdf5 exists with bindings in many languages This is offset by HFD5 being a single, C-based, strictly versioned API/ABI implementation deal, e.g. a browser based app can't ingest...

I'd argue that many of the most useful modules from SciPy in iOS programming are backed by C/C++ modules notably signal and image modules. The Fortran backed stuff is obviously...

https://apple.stackexchange.com/a/157502 describes cross compilation w/ Fortran support, might make this issue feasible

If the build for numpy were open sourced, we could probably adapt for SciPy, to lower work required for omz

that's a neat trick, to use f2c, but curious if that actually works wtih scipy.

> I’d love to know whether ios12 has brought us any closer to having scikit-learn on Pythonista, or the iPad in general. it's a toolchain/compiler problem, not a iOS version...

While I mentioned gradients, I think some feedback about how long model evaluations are taking would be helpful for any method, especially in Julia where wants to check that some...

I wrote the EulerMaruyama implementation and interested to update it (and include some improvements for non-centered form & a 2nd order scheme as well). It's not clear to me why...

@junpenglao thanks for the explanation. Indeed the port is much more complicated than I expected, so thanks for your work on that.

sorry to jump on here, (but I could neither join the forum by github or direct sign up (I never received the verification email)), I would like to not only...