Author: Maciej Fijalkowski <fij...@gmail.com> Branch: extradoc Changeset: r4089:22d1acba1fad Date: 2012-02-18 12:18 +0200 http://bitbucket.org/pypy/extradoc/changeset/22d1acba1fad/
Log: work some more on slides diff --git a/planning/micronumpy.txt b/planning/micronumpy.txt --- a/planning/micronumpy.txt +++ b/planning/micronumpy.txt @@ -42,3 +42,6 @@ - things like take/item/fancy indexing can use some knowledge about the density of data and either evaluate interesting points (without forcing) or do what they do now. + +- counting by element_size instead of by 1 and then multiply sounds + like a much faster option sometimes diff --git a/talk/sea2012/talk.rst b/talk/sea2012/talk.rst --- a/talk/sea2012/talk.rst +++ b/talk/sea2012/talk.rst @@ -5,8 +5,11 @@ ------------------------ * what is pypy and why + * numeric landscape in python + * what we achieved in pypy + * where we're going What is PyPy? @@ -52,7 +55,38 @@ Numerics in Python ------------------ -XXX numeric expressions, plots etc. +* ``numpy`` - for array operations + +* ``scipy``, ``scikits`` - various algorithms, also exposing C/fortran + libraries + +* ``matplotlib`` - pretty pictures + +* ``ipython`` + +There is an entire ecosystem! +----------------------------- + +* Which I don't even know very well + +* ``PyCUDA`` + +* ``pandas`` + +* ``mayavi`` + +What's important? +----------------- + +* There is an entire ecosystem built by people + +* It's available for free, no shady licensing + +* It's being expanded + +* It's growing + +* It'll keep up with hardware advancments Problems with numerics in python -------------------------------- @@ -92,12 +126,21 @@ * Assembler generation backend needs works -* No vectorization yet +* Vectorization in progress Status benchmarks ----------------- +* laplace solution + +* solutions: + + +---+ + | | + +---+ + This is just the beginning... ----------------------------- * PyPy is an easy platform to experiment with + _______________________________________________ pypy-commit mailing list pypy-commit@python.org http://mail.python.org/mailman/listinfo/pypy-commit