Skip to content

[Bugfix] Fix updating parameters under align_modules #361

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from
Closed

[Bugfix] Fix updating parameters under align_modules #361

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 1 addition & 19 deletions src/compressed_tensors/utils/offload.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,7 +78,6 @@
"delete_offload_parameter",
"has_offloaded_params",
"disable_hf_hook",
"disable_offload",
"align_modules",
"align_module_device",
"register_offload_module",
Expand Down Expand Up @@ -386,23 +385,6 @@ def delete_from_weights_map(
)


@check_accelerate(fallback=contextlib.nullcontext())
@contextlib.contextmanager
def disable_offload(module: torch.nn.Module):
"""
Context manager to disable module onloading and offloading. Parameters will stay on
their current device

:param module: module to disable offloading for
"""
if has_offloaded_params(module):
module._hf_hook.offload = False
yield
module._hf_hook.offload = True
else:
yield


@check_accelerate(fallback=contextlib.nullcontext())
@contextlib.contextmanager
def align_modules(
Expand All @@ -419,9 +401,9 @@ def align_modules(
modules = (modules,) if isinstance(modules, torch.nn.Module) else modules

with contextlib.ExitStack() as stack:
stack.enter_context(disable_offloading())
for module in modules:
stack.enter_context(align_module_device(module, execution_device))
stack.enter_context(disable_offload(module)) # disable redundant onloading
yield


Expand Down
8 changes: 6 additions & 2 deletions tests/test_utils/test_offload.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -305,6 +305,7 @@ def test_align_modules():
module0 = ExampleModule()
module1 = ExampleModule()
module2 = ExampleModule()
new_data = torch.tensor(1.0)
model = torch.nn.Sequential(module0, torch.nn.Sequential(module1, module2))
attach_align_device_hook(
model,
Expand All @@ -318,13 +319,16 @@ def test_align_modules():
assert module2.a.device == torch.device("meta")

with align_modules((module0, module1)):
assert module0.a.device != torch.device("meta")
assert module1.a.device != torch.device("meta")
assert module0.a.device == torch.device("cpu")
assert module1.a.device == torch.device("cpu")
assert module2.a.device == torch.device("meta")
update_offload_parameter(module0, "a", new_data)
assert module0.a == new_data

assert module0.a.device == torch.device("meta")
assert module1.a.device == torch.device("meta")
assert module2.a.device == torch.device("meta")
assert module0._hf_hook.weights_map["a"] == new_data


@requires_accelerate()
Expand Down