Saturday, January 28, 2012

NumPyPy status update

Hello.

This is just a quick status update on the NumPy in PyPy project that very recently became my day job. I should give my thanks once again to Getco, Nate Lawson and other contributors who donated above $40000 towards the goal.

Recently we (Alex Gaynor, Matti Picus and me) implemented a few interesting things that a lot of people use:

  • more ufuncs
  • most ufuncs now accept the axis parameter (except all and any)
  • fixed string representation of arrays, now it's identical to numpy (uses pretty much the same code)
  • ndarray.flat should be working correctly
  • ndarray.flatten, ndarray.ravel, ndarray.take
  • indexing arrays by boolean arrays of the same size
  • and various bugfixes.

We would also like to introduce the nightly report of numpy status. This is an automated tool that does package introspection. While it gives some sort of idea how much of numpy is implemented, it's not by far the authority. Your tests should be the authority. It won't report whether functions support all kinds of parameters (for example masked arrays and out parameter are completely unsupported) or that functions work at all. We also reserve the right to incorporate jokes in that website, so don't treat it that seriously overall :-)

Thanks, and stay tuned. We hope to post here regular updates on the progress.

Cheers,
fijal & the PyPy team

3 comments:

Anonymous said...

I use "out" parameter very often in my code (with numpy.take), without this one my code would run much worse (because huge arrays of hundreds MB would copy many times inside a big cycle). How currently the "out" parameter is handled (warning, error, nothing)?

Maciej Fijalkowski said...

It just errors with more or less acceptable error message. Note that pypy does not create intermediates for most of operations, so if you have a lot of them chained actually using out will be worse than not using it.

Anonymous said...

I'm new to python but not to Cpython/numpy/scipy/matplotlib and I fail to understand what you are doing.

* In a nutshell, what's numpypy? Is it a rewrite of the numpy code to make it compatible with pypy? or are you working on pypy itself to be able to run numpy as it is??

* if numpypy is a rewrite of numpy, that's good but how do you plan to keep numpy and numpypy sync (in terms of functionalities)??

* Using numpy with pypy will be great but what about scipy qnd matplotlib??
Many users need at least these two modules on top of numpy;

I would be very happy with pypy being able to work with unpachted numpy/scipy/matplotlib.

I think your website should summarise these issues on its front page.