You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

Frank Lee ae1b58cd16 [tensor] added linear implementation for the new sharding spec (#1416) ... * [tensor] added linear implementation for the new sharding spec * polish code		2 years ago
..
components_to_test	[test] ignore 8 gpu test (#1080)	3 years ago
test_amp	[test] refactored with the new rerun decorator (#763)	3 years ago
test_comm	[communication] add p2p_v2.py to support communication with List[Any] (#1407)	2 years ago
test_config	[pipeline] refactor the pipeline module (#1087)	3 years ago
test_context	[test] refactored with the new rerun decorator (#763)	3 years ago
test_data	[unittest] refactored unit tests for change in dependency (#838)	3 years ago
test_data_pipeline_tensor_parallel	polish unitest test with titans (#1152)	2 years ago
test_ddp	[zero] alleviate memory usage in ZeRODDP state_dict (#1398)	2 years ago
test_device	[device] add DeviceMesh class to support logical device layout (#1394)	2 years ago
test_engine	[hotfix] remove potiential circle import (#1307)	2 years ago
test_fx	[fx] fix the false interpretation of algorithm 3 in https://arxiv.org/abs/1604.06174. (#1446)	2 years ago
test_gemini	[zero] add chunk_managerV2 for all-gather chunk (#1441)	2 years ago
test_layers	[FAW] export FAW in _ops (#1438)	2 years ago
test_moe	[test] refactored with the new rerun decorator (#763)	3 years ago
test_ops	[FAW] export FAW in _ops (#1438)	2 years ago
test_optimizer	[hotfix] fix CPUAdam kernel nullptr (#1410)	2 years ago
test_pipeline	[pipeline] refactor the pipeline module (#1087)	3 years ago
test_tensor	[tensor] added linear implementation for the new sharding spec (#1416)	2 years ago
test_trainer	[pipeline] refactor the pipeline module (#1087)	3 years ago
test_utils	[utils] Impl clip_grad_norm for ColoTensor and ZeroOptimizer (#1442)	2 years ago
test_zero	[zero] zero optim state_dict takes only_rank_0 (#1384)	2 years ago
__init__.py	[zero] Update sharded model v2 using sharded param v2 (#323)	3 years ago