Permalink
Commits on Nov 13, 2018
  1. Corrected typo in README (#621)

    DaveVoyles authored and alsrgv committed Nov 13, 2018
Commits on Nov 3, 2018
Commits on Nov 1, 2018
  1. Use torch:: instead of at:: namespace in Torch backend (#596)

    goldsborough authored and alsrgv committed Nov 1, 2018
    * Use torch:: instead of at:: namespace
    
    * torch:: -> ::torch::
Commits on Oct 29, 2018
  1. PyTorch break: .dtype() -> .scalar_type() (#595)

    alsrgv committed Oct 29, 2018
    Details are in pytorch/pytorch#12766
Commits on Oct 25, 2018
  1. Bump version to 0.15.1 (#586)

    alsrgv committed Oct 25, 2018
Commits on Oct 24, 2018
  1. Added option to treat sparse gradients as dense tensors (#570)

    alsrgv committed Oct 24, 2018
Commits on Oct 23, 2018
  1. Add AVX intrinsics to vectorize & speed up FP16-CPU computations (#574)

    alsrgv committed Oct 23, 2018
    * Add AVX intrinsics to vectorize & speed up FP16-CPU computations
    
    * Add compilation test & runtime identification of AVX/F16C instinsics
Commits on Oct 17, 2018
  1. PyTorch master compatibility, improved errors in case of .so import f…

    alsrgv committed Oct 17, 2018
    …ailure (#569)
    
    * PyTorch master compatibility, improved errors in case of .so import failure
    
    * Temporarily add future dependency for PyTorch
    
    * Switch 0.4.2 (nonexistent) -> 1.0.0
Commits on Oct 16, 2018
  1. FP16 allreduce on CPU (#563)

    alsrgv committed Oct 16, 2018
Commits on Oct 15, 2018
  1. Broadcast optimizer options in addition to parameter state (#562)

    tgaddair committed Oct 15, 2018
    * Broadcast optimizer options in addition to parameter state
    
    * Added comment
    
    * Added tests for all the optimizer subclasses
    
    * Added comment
Commits on Oct 10, 2018
Commits on Oct 8, 2018
  1. Upgrade Dockerfile to TensorFlow 0.11.0, PyTorch 0.4.1 and Open MPI 3…

    alsrgv committed Oct 8, 2018
    ….1.2 (#544)
    
    * Upgrade Dockerfile to TensorFlow 0.11.0 and PyTorch 0.4.1
    
    * Upgrade Open MPI to 3.1.2
Commits on Oct 2, 2018
  1. Fully de-initialize Horovod upon shutdown (#532)

    alsrgv committed Oct 2, 2018
    * Fully de-initialize Horovod upon shutdown
    
    * Move clearing initialization flag to just before starting initialization
    
    * Add missing shut_down behaviors as well
Commits on Oct 1, 2018
Commits on Sep 29, 2018
Commits on Sep 28, 2018
  1. FP16 support for GPU tensors in all frameworks (#529)

    tgaddair committed Sep 28, 2018
    * Initial support for FP16
    
    Bump version to a dev release
    
    Cast vars to fp16 before allreduce to compress gradients
    
    Abstracted compression algorithm into a class hierarchy and added algorithm flag to optimizer and allreduce signatures
    
    Changed compressor to set the dtype on initialization
    
    Resolved conflicts
    
    Additional conflicts
    
    Formatting
    
    More formats
    
    Updated license
    
    Added fp16 compression for Keras
    
    Added arguments to keras examples
    
    Fixed imports
    
    * Added compression to tf.keras
    
    * Added PyTorch compression API
    
    Added unit tests
    
    Whitespace
    
    * Added C interfaces and types
    
    * Forward declare
    
    * Removed Half from older versions of PyTorch
    
    * Added error for old version of PyTorch
    
    * Removed reference to float16
    
    * Updated examples, added compression to the Keras model load
    
    * Cleaned imports
    
    * Removed dependency on enums
    
    * Updated unit tests
    
    * Test compatability fix
    
    * Reverted version updates
    
    * Fixed message
    
    * Removed imports
    
    * Added cuda.HalfTensor to all PyTorch tests with CUDA
    
    * Only compare versions once
    
    * Renamed --fp16 in examples to --fp16-allreduce for clarity
    
    * Replaced assignment with set_
    
    * Modified compression algorithms to be stateless with optional context parameters
    
    * Removed optional ctx parameter
    
    * Replaced 0.4.2 with 1.0.0
    
    * Only run GPU tests with HalfTensors if fp16 is supported
Commits on Sep 27, 2018
Commits on Sep 26, 2018
  1. Added tf.keras support (#513)

    tgaddair committed Sep 26, 2018
    * Added support for tf.keras
    
    * Added unit tests
    
    * Refactoring
    
    * Fixed tests
    
    * Hide implementation modules
    
    * Moved _DistributedOptimizer into the impl file and wrapped with function
    
    * Added cooperative multiple inheritance
    
    * Backwards compatability with TensorFlow versions less than 1.4.0
    
    * Removed duplicate headers
  2. PyTorch 1.0 readiness updates (#523)

    alsrgv committed Sep 26, 2018
    * PyTorch 1.0 readiness updates
    
    * Further updates, drop PyTorch 0.3.1 support
    
    * Add PyTorch version check
    
    * Switch to use LooseVersion for version comparison
Commits on Sep 18, 2018
Commits on Sep 14, 2018