Very clear, interesting post from @johnmyleswhite on Julia data frame performance: http://www.johnmyleswhite.com/notebook/2015/11/28/why-julias-dataframes-are-still-slow/ …
-
-
@johnmyleswhite though doesn't that sorta back away from the design principle of "it's ok to work with loops and scalars"? -
@avibryant There's currently too much essential type-information that's available, but underutilized, because of the eng costs.
End of conversation
New conversation -
-
-
@avibryant@johnmyleswhite there has been some design exploration in Haskell by@cartazio (typed matrices) and@a_cowley (typed dataframes) -
@nathanhowell@avibryant@johnmyleswhite@cartazio And we're going to get andrewthad's join relational stuff to work with Frames! -
@a_cowley@nathanhowell@avibryant@johnmyleswhite was that at hac phi? Linkeee? -
@cartazio@nathanhowell@avibryant@johnmyleswhite Yes! Convo, blackboard, and mutual demo. But we got onto the same page. -
@a_cowley@nathanhowell@avibryant@johnmyleswhite cool plz share links when you have a chance :) -
@cartazio@nathanhowell@avibryant@johnmyleswhite I will! We have top notch perf, and Andrew's join work is an awesome API.
End of conversation
New conversation -
-
-
@avibryant@johnmyleswhite : but you'll never get away from index access for values entirely. V8 specializes some arrays as a result. -
@slightlylate@avibryant Is there any documentation on this? Would be interested to understand more. -
@johnmyleswhite there's no good V8 documentation but it's similar to the PyPy approach http://tratt.net/laurie/research/pubs/html/bolz_diekmann_tratt__storage_strategies_for_collections_in_dynamically_typed_languages/ …@slightlylate@avibryant -
@mraleph@slightlylate@avibryant This is great. Thank you!
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.