1. 17 Sep, 2020 1 commit
    • Tom Birch's avatar
      Multi-process pipe (#90) · 63f7796a
      Tom Birch authored
      Adds support for distributing pipeline stages across multiple processes (and therefore multiple machines)
      * Adds a style argument to the Pipe constructor, defaulting to PipelineStyle.SingleProcess, but also supporting PipelineStyle.MultiProcess
      * Added support for lazy construction of modules (see lazy_construction for an example)
      * Added two implementations of inter-process communication: one based on rpc with globally visible queues, one based on send/recv
      * Copied all the relevant tests from tests/pipe to tests/pipe_process and modified them to exercise PipelineStyle.MultiProcess
      63f7796a
  2. 16 Sep, 2020 1 commit
  3. 03 Sep, 2020 1 commit
    • Jun Ru Anderson's avatar
      Add grad scaler (#48) · b6a5e634
      Jun Ru Anderson authored
      
      
      Add GradScaler to Fairscale, subclassing PyTorch's GradScaler. Use GradScaler in the pipe benchmark; though it is not needed in this case, it is a good example of how to use gradient scaling for larger models that do require gradient scaling in order to converge.
      Co-authored-by: default avatarJun Ru Anderson <andersonic@fb.com>
      b6a5e634
  4. 27 Aug, 2020 1 commit
  5. 14 Aug, 2020 2 commits
  6. 31 Jul, 2020 3 commits
  7. 08 Jul, 2020 1 commit