Skip to content
This repository has been archived by the owner on Nov 17, 2023. It is now read-only.

fix potential floating number overflow, enable float16 #12118

Merged
merged 14 commits into from
Aug 20, 2018
Merged

fix potential floating number overflow, enable float16 #12118

merged 14 commits into from
Aug 20, 2018

Commits on Aug 16, 2018

  1. fix potential floating number overflow, enable float16

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    30d85e2 View commit details
    Browse the repository at this point in the history

  2. fix cuda impl

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    9daf3f5 View commit details
    Browse the repository at this point in the history

  3. fix cuda imple

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    edb31e8 View commit details
    Browse the repository at this point in the history

  4. fix template substitution for windows

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    0ab0695 View commit details
    Browse the repository at this point in the history

  5. half_f substantiate operand + fix

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    0b90e5a View commit details
    Browse the repository at this point in the history

  6. remove ambiguous operand + for mshadow half_T

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    6915172 View commit details
    Browse the repository at this point in the history

  7. fix con't

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    62a106e View commit details
    Browse the repository at this point in the history

  8. use int32_t as indices

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    4c86438 View commit details
    Browse the repository at this point in the history

  9. use overload

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    565a9c1 View commit details
    Browse the repository at this point in the history

  10. try remove ambiguous function overloading

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    38eeece View commit details
    Browse the repository at this point in the history

  11. thrust version limit

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    75c9240 View commit details
    Browse the repository at this point in the history

  12. change sizeof cast from floor to ceil when allocating buffers

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    c5688c1 View commit details
    Browse the repository at this point in the history

  13. cleaner

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    e72707c View commit details
    Browse the repository at this point in the history

  14. fix alignment of pointers

    @zhreshold
    zhreshold committed Aug 16, 2018
    Configuration menu
    Copy the full SHA
    e09ef1a View commit details
    Browse the repository at this point in the history