Skip to content
Snippets Groups Projects
  1. May 05, 2022
  2. Apr 27, 2022
    • Zaida Zhou's avatar
      [Enhancement] Handle the device type of inputs in functions (#137) · 17dbac18
      Zaida Zhou authored
      * [Enhancement] Handle the device type of inputs in functions
      
      * rename and move three fucntions to dist/utils.py
      
      * minor refinement
      
      * rename dist to torch_dist in utils.py
      
      * update unit tests
      
      * refine unit tests
      
      * add unit tests
      
      * fix unit tests
      
      * replace Sequence with list and tuple
      
      * rename get_backend_device to get_comm_device
      
      * fix unit tests
      
      * fix unit tests
      
      * refactor and add more unit tests
      
      * cast_data_device does not support set type
  3. Mar 05, 2022
    • Zaida Zhou's avatar
      [Feature] Add distributed module (#59) · c6a8d72c
      Zaida Zhou authored
      * [Feature] Add distributed module
      
      * fix IS_DIST error
      
      * all_reduce_dict does operations in-place
      
      * support 'mean' operation
      
      * provide local group process
      
      * add tmpdir argument for collect_results
      
      * add unit tests
      
      * refactor unit tests
      
      * simplify steps to create multiple processes
      
      * minor fix
      
      * describe the different of *gather* in mmengine and pytorch
      
      * minor fix
      
      * add unit tests for nccl
      
      * test nccl backend in multiple gpu
      
      * add get_default_group function to handle different torch versions
      
      * minor fix
      
      * [Feature] Add distributed module
      
      * fix IS_DIST error
      
      * all_reduce_dict does operations in-place
      
      * support 'mean' operation
      
      * provide local group process
      
      * add tmpdir argument for collect_results
      
      * add unit tests
      
      * refactor unit tests
      
      * simplify steps to create multiple processes
      
      * minor fix
      
      * describe the different of *gather* in mmengine and pytorch
      
      * minor fix
      
      * add unit tests for nccl
      
      * test nccl backend in multiple gpu
      
      * add get_default_group function to handle different torch versions
      
      * minor fix
      
      * minor fix
      
      * handle torch1.5
      
      * handle torch1.5
      
      * minor fix
      
      * fix typo
      
      * refactor unit tests
      
      * nccl does not support gather and gather_object
      
      * fix gather
      
      * fix collect_results_cpu
      
      * fix collect_results and refactor unit tests
      
      * fix collect_results unit tests
      
      * handle torch.cat in torch1.5
      
      * refine docstring
      
      * refine docstring
      
      * fix comments
      
      * fix comments
Loading