Live data from Hacker News

Viewing profile — tadlan

tadlan

HN member
Joined
Thu, Nov 20, 2014, 7:20 AM UTC
HN karma
26
Public activity
34 items

About tadlan

No profile information was provided.

Recent public activity

  1. comment
  2. comment
    Comment #11290509

    Thanks. I'm not familiar with dynamic linking. What will that do?

  3. comment
    Comment #11290260

    Any specific plans for python support?

  4. comment
    Comment #11290196

    Or julia. Which is 1000th better than js

  5. comment
    Comment #11290188

    It will eventually, iirc

  6. comment
    Comment #11290042

    Question please: Even though there is stated desire to support python etc vm, this will be hampered by the need for the client to download the large VM vs native code that can be d…

  7. comment
    Comment #11237708

    Have you tried dask?

  8. comment
    Comment #11019424

    What about dynd? Iirc doesn't use iteration st all.

  9. comment
    Comment #11004681

    Woah!!! Cool

  10. comment
    Comment #10830191

    Make sure you use the devect macro or parallel accelerator if you're going to use vector iced code :)

  11. comment
    Comment #10830114

    It's not a would and gpod. Numba works like that right now, has better interoperability with python and less context switching. Numba is I easily installed on all systems through c…

  12. comment
    Comment #10830109

    Actually, Numba array allocation in nopython mode is allowed now. What other numerical features are missing?

  13. comment
    Comment #10830104

    Numba obviate these issues

  14. comment
    Comment #10829507

    What would one need to use c or c++ to write a julia package? It's as fast as native code so no need to use multiple languages.

  15. comment
    Comment #10829038

    Numba would still be faster. It would fuse away any intermediates in the code and remove any Overhead to the compiled code. also have the option of devecting to loops. both of whic…

  16. comment
    Comment #10828823

    Numbs compiles entire functions on and allows array expressions with allocation and loop fusion. I don't see the problem

  17. comment
    Comment #10828663

    Use numba to compile python loops or array expressions to fast llvm, and problem solved. I'm sticking with python.

  18. comment
  19. comment
    Comment #10644825

    That's just the data structure. Fitting models would be much tougher.

  20. comment
    Comment #10569650

    Will julia ever be popular for general computing and Web stuff? It's s till fast for non scientific use cases, has multi threading (in master) and is getting go like channels.

  21. comment
    Comment #10485202

    Julia!!!!! The answer to all these problems, both for general and science programming.

  22. comment
    Comment #10389798

    Dask has that, and scikit learn is moving that way also. It even beats spark for out of core work on a single machine

  23. comment
    Comment #10389716

    There's a dev branch that has been actively developed

  24. comment
    Comment #10270023

    Numba has list, Named tuples, and is working on define your own jit classes.

  25. comment
    Comment #10270020

    Numba has list, Named tuples, and is working on define your own jit classes.