Skip to content

SMP PT 2.3 Fix #4796

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

Merged
merged 12 commits into from
Jul 26, 2024
Merged
1 change: 0 additions & 1 deletion src/sagemaker/fw_utils.py
Original file line number Diff line number Diff line change
Expand Up @@ -152,7 +152,6 @@
"2.1.0",
"2.1.2",
"2.2.0",
"2.3.0",
"2.3.1",
]

Expand Down
2 changes: 1 addition & 1 deletion src/sagemaker/image_uri_config/pytorch-smp.json
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@
"2.1": "2.1.2",
"2.2": "2.3.1",
"2.2.0": "2.3.1",
"2.3": "2.4.0"
"2.3.1": "2.4.0"
},
"versions": {
"2.0.1": {
Expand Down
6 changes: 6 additions & 0 deletions tests/unit/sagemaker/image_uris/test_smp_v2.py
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,7 @@ def test_smp_v2(load_config):
"torch_distributed": {"enabled": True},
"smdistributed": {"modelparallel": {"enabled": True}},
}

for processor in PROCESSORS:
for version in VERSIONS:
ACCOUNTS = load_config["training"]["versions"][version]["registries"]
Expand All @@ -38,6 +39,11 @@ def test_smp_v2(load_config):
if "2.1" in version or "2.2" in version or "2.3" in version:
cuda_vers = "cu121"

if "2.3.1" == version:
py_version = "py311"

print(version, py_version)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Remove this print

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Oops, accidentally left that in. Removing now

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Or wait, it is merged already, is it worth making a new PR for it?


uri = image_uris.get_training_image_uri(
region,
framework="pytorch",
Expand Down