OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-07 12:21:27 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity
381d0cb239
pytorch/test/distributed/tensor
History
Tugsbayasgalan Manlaibaatar 6b1b95ad2a Support subclass constructor capturing in export (#147014) 
Notable TODOs:
1. Need to implement AutogradHOP to get rid of subclasses before serializing
2. Need to implement mechanism to figure out what subclasses will be used in export when they are not expressed in the inputs

Differential Revision: [D69640673](https://our.internmc.facebook.com/intern/diff/D69640673)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/147014
Approved by: https://github.com/bdhirsh
2025-03-16 18:19:19 +00:00
..
debug PEP585 update - test (#145176) 2025-01-22 04:48:28 +00:00
experimental PEP585 update - test (#145176) 2025-01-22 04:48:28 +00:00
parallel [async TP] insert reshape node to handle "reshape -> scaled mm -> reshape pattern" in async TP with rowwise scales (#148001) 2025-03-02 03:25:28 +00:00
__init__.py [dtensor] move all tests to distribute/tensor folder (#144166) 2025-01-08 00:32:33 +00:00
README.md [dtensor] move all tests to distribute/tensor folder (#144166) 2025-01-08 00:32:33 +00:00
test_api.py [TP] Add warning when module is distributed twice (#147006) 2025-02-13 06:49:17 +00:00
test_attention.py [dtensor] add CuDNN SDPA op support to DTensor (#148537) 2025-03-06 23:44:40 +00:00
test_common_rules.py [dtensor] move all tests to distribute/tensor folder (#144166) 2025-01-08 00:32:33 +00:00
test_convolution_ops.py [DTensor][conv] add DTensor convolution_backward op support for case where the input Tensor has requires_grad=False (#142278) 2025-02-10 07:06:40 +00:00
test_dtensor_compile.py Support subclass constructor capturing in export (#147014) 2025-03-16 18:19:19 +00:00
test_dtensor_ops.py [DTensor] Add pointwise ops strategy for aten.minimum (#145816) 2025-01-29 01:19:01 +00:00
test_dtensor.py [dtensor] refactor sharding prop to handle cross mesh computation (#147869) 2025-03-04 18:30:44 +00:00
test_embedding_ops.py [dtensor] move all tests to distribute/tensor folder (#144166) 2025-01-08 00:32:33 +00:00
test_experimental_ops.py [dtensor] move all tests to distribute/tensor folder (#144166) 2025-01-08 00:32:33 +00:00
test_init.py [dtensor] move all tests to distribute/tensor folder (#144166) 2025-01-08 00:32:33 +00:00
test_math_ops.py [dtensor] refactor sharding prop to handle cross mesh computation (#147869) 2025-03-04 18:30:44 +00:00
test_matrix_ops.py [DTensor][Test] Create a simple unit test for tensordot (#146514) 2025-02-11 21:57:56 +00:00
test_op_strategy.py [dtensor] refactor sharding prop to handle cross mesh computation (#147869) 2025-03-04 18:30:44 +00:00
test_optimizers.py [dtensor] move all tests to distribute/tensor folder (#144166) 2025-01-08 00:32:33 +00:00
test_pointwise_ops.py PEP585 update - test (#145176) 2025-01-22 04:48:28 +00:00
test_random_ops.py [DTensor][random] defer DTensor RNG state sync until first random op call or manual_seed call; support more flexible OffsetBasedRNGTracker init (#147025) 2025-02-26 17:33:22 +00:00
test_redistribute.py Tests Generelization for multiple accelerator devices (#139749) 2025-01-14 08:52:46 +00:00
test_tensor_ops.py [dtensor] fix side-effect on dtype for _like ops (#146869) 2025-02-12 08:42:14 +00:00
test_utils.py [dtensor] move all tests to distribute/tensor folder (#144166) 2025-01-08 00:32:33 +00:00
test_view_ops.py [DTensor][Test] Add a test to demonstrate current dtensor view behavior if redistribution happens (#148015) 2025-03-01 10:24:40 +00:00
test_xla_integration.py PEP585 update - test (#145176) 2025-01-22 04:48:28 +00:00

README.md

Run distributed tensor tests:

from root, run (either CPU or GPU)

pytest test/distributed/tensor/test_dtensor.py

run specific test cases and print stdout/stderr:

pytest test/distributed/tensor/test_dtensor.py -s -k test_from_local