wz337 bb67a28738 [DTensor] Enable Adamax foreach optimizer (#119850) ... Enable Adamax foreach optimizer and add DTensor unit test for Adamax. Pull Request resolved: https://github.com/pytorch/pytorch/pull/119850 Approved by: https://github.com/wanchaol		2024-02-14 20:43:00 +00:00
..
_composable	[FSDP2] Added mixed precision (#118223)	2024-02-13 19:05:30 +00:00
_shard
_spmd	Update to TorchFix 0.4.0 (#119424)	2024-02-12 23:30:12 +00:00
_tensor	[DTensor] Enable Adamax foreach optimizer (#119850)	2024-02-14 20:43:00 +00:00
_tools
algorithms
bin
checkpoint	[DCP] Enable filesystem/fsspec auto detection (#118888)	2024-02-08 16:38:04 +00:00
elastic	[TorchElastic] Handle SystemExit with code == 0 (#119697)	2024-02-14 03:09:09 +00:00
fsdp	Remove parent device mesh check (#118620)	2024-02-08 00:49:28 +00:00
launcher
nn/jit
optim
pipeline/sync
rpc
tensor/parallel	add torch.float64 precision support to the transformer test suite in TP/SP (#116436)	2024-01-30 23:50:29 +00:00
argparse_util_test.py
test_c10d_common.py
test_c10d_functional_native.py	Fix some tests in test_c10d_functional_native.py (#119102)	2024-02-12 09:28:18 +00:00
test_c10d_gloo.py	ProcessGroupGloo::reduce_scatter_tensor_coalesced (#118911)	2024-02-03 02:42:47 +00:00
test_c10d_logger.py
test_c10d_nccl.py	[c10d][flight recorder] store a copy of string in entry (#119837)	2024-02-14 09:13:56 +00:00
test_c10d_object_collectives.py
test_c10d_pypg.py
test_c10d_spawn.py
test_c10d_spawn_gloo.py
test_c10d_spawn_nccl.py
test_c10d_spawn_ucc.py
test_c10d_ucc.py
test_collective_utils.py
test_compute_comm_reordering.py	[reland] Fix estimate_nccl_collective_runtime (#118986)	2024-02-12 18:48:06 +00:00
test_data_parallel.py
test_device_mesh.py	[DeviceMesh][Test] Remove test_raises_mesh_dim_less_than_2 (#119172)	2024-02-05 17:34:51 +00:00
test_distributed_spawn.py
test_dynamo_distributed.py	Fix failure of test_dynamo_distributed & test_inductor_collectives (#117741)	2024-01-25 13:25:36 +00:00
test_fake_pg.py
test_functional_api.py	[Distributed] Fix tests when CUDA not available (#117163)	2024-01-11 22:27:43 +00:00
test_inductor_collectives.py	[inductor] Slightly faster memory allocation on CUDA (#118255)	2024-01-25 20:49:14 +00:00
test_launcher.py
test_multi_threaded_pg.py
test_nccl.py
test_pg_wrapper.py
test_store.py	[c10d] Add a recursive method to get the inner most store (#117074)	2024-01-10 20:22:55 +00:00