| .. |
|
_composable
|
[Compile FSDP2][3/n] Check all_gather_work is distributed_c10d.Work before calling .wait() (#123491)
|
2024-04-10 01:23:54 +00:00 |
|
_shard
|
[PT] [ST] support non contiguous rank validation in sharded tensor (#123230)
|
2024-04-05 21:05:01 +00:00 |
|
_sharded_tensor
|
|
|
|
_sharding_spec
|
|
|
|
_spmd
|
[dtensor] refactor schema suggestions in output sharding (#122929)
|
2024-04-01 17:39:39 +00:00 |
|
_tensor
|
DTensor: add ring attention for _scaled_dot_product_flash_attention (#122460)
|
2024-04-03 06:45:00 +00:00 |
|
_tools
|
|
|
|
algorithms
|
[DDP] Use compiled_autograd to trace DDP backward allreduce (#110662)
|
2024-02-08 03:03:15 +00:00 |
|
autograd
|
|
|
|
benchmarks
|
Move doc links to point to main (#121823)
|
2024-03-15 19:49:37 +00:00 |
|
checkpoint
|
[DCP] Adds better handling in logging of specific kwargs (#123658)
|
2024-04-11 21:09:38 +00:00 |
|
elastic
|
Adding health check server hook in torch elastic (#122750) (#123504)
|
2024-04-11 19:10:56 +00:00 |
|
examples
|
|
|
|
fsdp
|
[FSDP1] fix _same_storage check for DTensor (#123617)
|
2024-04-10 10:26:12 +00:00 |
|
launcher
|
[Torchelastic][Logging] Pluggable logsspecs using python entrypoints and option to specify one by name. (#120942)
|
2024-03-02 08:07:52 +00:00 |
|
nn
|
Fix get_rank under a non-default group. (#120481)
|
2024-03-11 05:40:54 +00:00 |
|
optim
|
Add tensor step and capturable support to rprop (#122261)
|
2024-03-28 23:31:18 +00:00 |
|
pipeline
|
[c10d] Deprecate torch.distributed.pipeline (#121464)
|
2024-03-08 19:55:02 +00:00 |
|
rpc
|
|
|
|
tensor
|
[TP] Add wildcard support (#122968)
|
2024-04-02 21:23:39 +00:00 |
|
__init__.py
|
|
|
|
_composable_state.py
|
|
|
|
_functional_collectives_impl.py
|
[funcol] add deprecation warning for the legacy backend (#122666)
|
2024-04-03 00:27:06 +00:00 |
|
_functional_collectives.py
|
Don't create world pg variable out of thin air when rewriting c10d collectives (#122561)
|
2024-03-26 20:12:08 +00:00 |
|
_state_dict_utils.py
|
[DCP] Adds ability to create a CPU state dict that is both shared and pinned (#122338)
|
2024-04-03 20:05:01 +00:00 |
|
argparse_util.py
|
Add --local-ranks-filter to torchrun: allow logs filtering by rank (#118562)
|
2024-02-07 04:29:54 +00:00 |
|
c10d_logger.py
|
[DCP] Adds better handling in logging of specific kwargs (#123658)
|
2024-04-11 21:09:38 +00:00 |
|
collective_utils.py
|
|
|
|
constants.py
|
|
|
|
CONTRIBUTING.md
|
|
|
|
device_mesh.py
|
[DeviceMesh] Make dtype of mesh tensor from init_device_mesh() consistent with directly calling DeviceMesh() (#123677)
|
2024-04-10 09:14:34 +00:00 |
|
distributed_c10d.py
|
Fix example in torch.distributed.new_subgroups docstring (#123492)
|
2024-04-10 03:33:07 +00:00 |
|
launch.py
|
[BE] minor logging cleanup in distributed (#122921)
|
2024-03-29 03:34:01 +00:00 |
|
logging_handlers.py
|
|
|
|
remote_device.py
|
|
|
|
rendezvous.py
|
Enable local_partial_types (#118467)
|
2024-01-28 13:38:22 +00:00 |
|
run.py
|
[BE] minor logging cleanup in distributed (#122921)
|
2024-03-29 03:34:01 +00:00 |
|
utils.py
|
|
|
13070e2753