saymrwulf/pytorch
1
0
Fork 0
mirror of https://github.com/saymrwulf/pytorch.git synced 2026-05-14 20:57:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
pytorch/test/distributed/_tensor
History
Anshul Sinha 83617017e0 [dtensor][debug] add c10d allreduce_coalesced_ tracing to CommDebugMode (#127040) 
**Summary**
Added c10d all_reduce_coalesced tracing to CommDebugMode and added test case to test_comm_mode.py.

**Test Plan**
pytest test/distributed/_tensor/debug/test_comm_mode.py

Pull Request resolved: https://github.com/pytorch/pytorch/pull/127040
Approved by: https://github.com/XilunWu
ghstack dependencies: #127025, #127029
2024-05-24 22:25:44 +00:00
..
debug [dtensor][debug] add c10d allreduce_coalesced_ tracing to CommDebugMode (#127040) 2024-05-24 22:25:44 +00:00
experimental [RELAND] Switch default behavoir of export IR to be predispatch (#125860) 2024-05-10 17:36:53 +00:00
__init__.py
README.md
test_api.py Revert "try to fix the warning in distribute_tensor (#125476)" 2024-05-04 22:25:32 +00:00
test_attention.py
test_common_rules.py
test_convolution_ops.py
test_dtensor.py [dtensor] early return for _split_tensor (#125810) 2024-05-14 04:59:27 +00:00
test_dtensor_compile.py AsyncCollectiveTensor: prevent wait_tensor() calls on graph inputs from getting DCEd (#125677) 2024-05-08 15:54:01 +00:00
test_dtensor_ops.py [DTensor] Supported simple replicate strategy for SVD (#127004) 2024-05-24 04:34:43 +00:00
test_embedding_ops.py
test_experimental_ops.py
test_init.py
test_math_ops.py [DTensor] Supported simple replicate strategy for SVD (#127004) 2024-05-24 04:34:43 +00:00
test_matrix_ops.py [dtensor] add op support for memory efficient attention (#122996) 2024-05-08 17:08:27 +00:00
test_op_strategy.py
test_optimizers.py [DTensor][Optim] Add support for fused_adam and fused_adamw when lr is a tensor (#126750) 2024-05-21 21:38:05 +00:00
test_pointwise_ops.py
test_random_ops.py
test_redistribute.py
test_tensor_ops.py [dtensor] improve new factory strategy (#122995) 2024-05-08 17:05:07 +00:00
test_utils.py
test_view_ops.py [dtensor] refactor view ops to use OpStrategy (#126011) 2024-05-17 05:39:21 +00:00
test_xla_integration.py

README.md

Run distributed tensor tests:

from root, run (either CPU or GPU)

pytest test/spmd/tensor/test_tensor.py

pytest test/spmd/tensor/test_ddp.py

run specific test case and print stdout/stderr:

pytest test/spmd/tensor/test_tensor.py -s -k test_tensor_from_local