Skip to content
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.

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

Add OOM observer with memory visualizations #2958

Merged
merged 34 commits into from
Feb 2, 2024
Merged
Changes from 1 commit
Commits
Show all changes
34 commits
Select commit Hold shift + click to select a range
bb75f65
add oomobserver
cli99 Feb 1, 2024
0054b52
update docstring
cli99 Feb 1, 2024
2e96d9c
Merge branch 'dev' into oom-observer
cli99 Feb 1, 2024
92faf4f
Update composer/callbacks/oom_observer.py
cli99 Feb 1, 2024
a0c6696
use pyskip
cli99 Feb 1, 2024
99395a8
call trainer fit
cli99 Feb 1, 2024
875546a
fix ci
cli99 Feb 1, 2024
66d093c
Update composer/callbacks/oom_observer.py
cli99 Feb 2, 2024
f2f94d3
addresss comments
cli99 Feb 2, 2024
43118ca
Merge branch 'dev' into oom-observer
cli99 Feb 2, 2024
5a74d34
Update composer/callbacks/oom_observer.py
cli99 Feb 2, 2024
ba6c859
add test wiht snapshot
cli99 Feb 2, 2024
637208d
update doc
cli99 Feb 2, 2024
cc23887
fix typo
cli99 Feb 2, 2024
c314a5c
use log info
cli99 Feb 2, 2024
1d0553b
fix format
cli99 Feb 2, 2024
1f2bf43
fix format
cli99 Feb 2, 2024
95104b5
fix ci
cli99 Feb 2, 2024
1faf75b
fix cpu test
cli99 Feb 2, 2024
bddca6c
Merge branch 'dev' into oom-observer
cli99 Feb 2, 2024
9d4e02d
fix ci
cli99 Feb 2, 2024
1e8c98f
Update tests/callbacks/test_oom_observer.py
cli99 Feb 2, 2024
f5d6db7
Update composer/callbacks/oom_observer.py
cli99 Feb 2, 2024
b48a720
Update composer/callbacks/oom_observer.py
cli99 Feb 2, 2024
b860bc0
Update composer/callbacks/oom_observer.py
cli99 Feb 2, 2024
07a8bec
update test
cli99 Feb 2, 2024
f91b854
Update tests/callbacks/test_oom_observer.py
cli99 Feb 2, 2024
7b7f30c
Update tests/callbacks/test_oom_observer.py
cli99 Feb 2, 2024
78bce44
Update tests/callbacks/test_oom_observer.py
cli99 Feb 2, 2024
818f772
Update composer/callbacks/oom_observer.py
cli99 Feb 2, 2024
c0ca7aa
Update composer/callbacks/oom_observer.py
cli99 Feb 2, 2024
5a07ae4
use warnings
cli99 Feb 2, 2024
74c66ce
add pytest filter user warnings in cpu callback tests
cli99 Feb 2, 2024
fe3dd2c
fix typo
cli99 Feb 2, 2024
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
Prev Previous commit
Next Next commit
Update composer/callbacks/oom_observer.py
Co-authored-by: Charles Tang <j316chuck@users.noreply.github.com>
cli99 and j316chuck authored Feb 2, 2024
commit 66d093cd7da5648fd762d5423b7ba433d7f805c7
2 changes: 1 addition & 1 deletion composer/callbacks/oom_observer.py
Original file line number Diff line number Diff line change
@@ -91,7 +91,7 @@ def __init__(
self._enabled = True
else:
self._enabled = False
log.warning('OOMObserver is supported after PyTorch 2.1.0. Skipping oom observer callback.')
warnings.warn('OOMObserver is supported after PyTorch 2.1.0. Skipping oom observer callback.')

def init(self, state: State, logger: Logger) -> None:
if not self._enabled: