DeepSpeed
e0a9e5a0 - Merge branch 'master' into olruwase/fast_persist

Comment changes are shownComment changes are hidden
Commit
41 days ago
Merge branch 'master' into olruwase/fast_persist
  • .github/workflows
    • File
      nv-torch-latest-v100.yml
    • setup-venv
      • File
        action.yml
    • File
      xpu-max1100.yml
  • File
    COMMITTERS.md
  • accelerator
    • File
      real_accelerator.py
    • File
      xpu_accelerator.py
  • deepspeed
    • module_inject
      • File
        auto_tp.py
      • File
        layers.py
    • runtime
      • swap_tensor
        • File
          __init__.py
        • File
          optimizer_utils.py
        • File
          partitioned_optimizer_swapper.py
        • File
          pipelined_optimizer_swapper.py
        • File
          utils.py
      • zero
        • File
          offload_states.py
        • File
          parameter_offload.py
        • File
          partition_parameters.py
        • File
          partitioned_param_coordinator.py
        • File
          stage3.py
    • utils
      • File
        __init__.py
      • File
        numa.py
      • File
        tensor_fragment.py
  • docs/code-docs/source
    • File
      zero3.rst
  • op_builder
    • File
      async_io.py
    • cpu
      • File
        async_io.py
    • npu
      • File
        async_io.py
    • xpu
      • File
        async_io.py
  • scripts
    • File
      check-extraindexurl.py
    • File
      check-license.py
  • tests/unit
    • inference/quantization
      • File
        test_intX_quantization.py
    • runtime
      • compile
        • File
          test_compile_zero.py
      • zero
        • File
          test_offload_states.py
        • File
          test_zero_multiple_run.py
        • File
          test_zero_tensor_fragment.py
  • File
    version.txt