1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
|
# Owner(s): ["oncall: distributed"]
import sys
from torch import distributed as dist
from torch.distributed.fsdp import FullyShardedDataParallel as FSDP
from torch.testing._internal.common_distributed import skip_if_lt_x_gpu
from torch.testing._internal.common_fsdp import (
CUDAInitMode,
FSDPInitMode,
FSDPTest,
NestedWrappedModule,
)
from torch.testing._internal.common_utils import (
TEST_WITH_DEV_DBG_ASAN,
run_tests,
)
if not dist.is_available():
print("Distributed not available, skipping tests", file=sys.stderr)
sys.exit(0)
if TEST_WITH_DEV_DBG_ASAN:
print(
"Skip dev-asan as torch + multiprocessing spawn have known issues",
file=sys.stderr,
)
sys.exit(0)
class TestTraversal(FSDPTest):
@property
def world_size(self):
return 2
@skip_if_lt_x_gpu(2)
def test_fsdp_modules(self):
nested_wrapped_module = NestedWrappedModule.init(
self.process_group,
FSDPInitMode.RECURSIVE,
CUDAInitMode.CUDA_BEFORE,
)
modules = FSDP.fsdp_modules(nested_wrapped_module)
self.assertEquals(
modules, [
nested_wrapped_module.module.get_submodule("1"),
nested_wrapped_module.module.get_submodule("1").get_submodule("0"),
nested_wrapped_module.module.get_submodule("2"),
]
)
modules = FSDP.fsdp_modules(nested_wrapped_module, root_only=True)
self.assertEqual(
modules, [
nested_wrapped_module.module.get_submodule("1"),
nested_wrapped_module.module.get_submodule("2"),
]
)
if __name__ == "__main__":
run_tests()
|
