Making large AI models cheaper, faster and more accessible
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.
 
 
 
 
 
YuliangLiu0306 1b494ad73c
[autoparallel] fix linear logical convert issue (#1857)
2 years ago
..
components_to_test [NFC] polish test component gpt code style (#1567) 2 years ago
test_amp [amp] add torch amp test (#1860) 2 years ago
test_auto_parallel [autoparallel] fix linear logical convert issue (#1857) 2 years ago
test_comm
test_config
test_context
test_data
test_data_pipeline_tensor_parallel [engin/schedule] use p2p_v2 to recontruct pipeline_schedule (#1408) 2 years ago
test_ddp [zero] add chunk init function for users (#1729) 2 years ago
test_device [tensor] support runtime ShardingSpec apply (#1453) 2 years ago
test_engine
test_fx [inference] streaming Linear 1D Row inference (#1874) 2 years ago
test_gemini [hotfix] fix zero's incompatibility with checkpoint in torch-1.12 (#1786) 2 years ago
test_layers [inference] streaming Linear 1D Row inference (#1874) 2 years ago
test_moe [moe] initialize MoE groups by ProcessGroup (#1640) 2 years ago
test_ops
test_optimizer
test_pipeline [fx/meta/rpc] move _meta_registration.py to fx folder / register fx functions with compatibility checks / remove color debug (#1710) 2 years ago
test_tensor [utils] remove lazy_memory_allocate from ColoInitContext (#1844) 2 years ago
test_trainer
test_utils [CheckpointIO] a uniform checkpoint I/O module (#1689) 2 years ago
test_zero [feature] A new ZeRO implementation (#1644) 2 years ago
__init__.py