Skip to content

[SMP] Changes for PT 2.4 currency upgrade #4890

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

Merged
merged 1 commit into from
Oct 15, 2024
Merged

[SMP] Changes for PT 2.4 currency upgrade #4890

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
1 change: 1 addition & 0 deletions src/sagemaker/fw_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -153,6 +153,7 @@
"2.1.2",
"2.2.0",
"2.3.1",
"2.4.1",
]

TRAINIUM_SUPPORTED_DISTRIBUTION_STRATEGIES = ["torch_distributed"]
Expand Down
28 changes: 27 additions & 1 deletion src/sagemaker/image_uri_config/pytorch-smp.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,8 @@
"2.1": "2.1.2",
"2.2": "2.3.1",
"2.2.0": "2.3.1",
"2.3.1": "2.5.0"
"2.3.1": "2.5.0",
"2.4.1": "2.6.0"
},
"versions": {
"2.0.1": {
Expand Down Expand Up @@ -160,6 +161,31 @@
"us-west-2": "658645717510"
},
"repository": "smdistributed-modelparallel"
},
"2.6.0": {
"py_versions": [
"py311"
],
"registries": {
"ap-northeast-1": "658645717510",
"ap-northeast-2": "658645717510",
"ap-northeast-3": "658645717510",
"ap-south-1": "658645717510",
"ap-southeast-1": "658645717510",
"ap-southeast-2": "658645717510",
"ca-central-1": "658645717510",
"eu-central-1": "658645717510",
"eu-north-1": "658645717510",
"eu-west-1": "658645717510",
"eu-west-2": "658645717510",
"eu-west-3": "658645717510",
"sa-east-1": "658645717510",
"us-east-1": "658645717510",
"us-east-2": "658645717510",
"us-west-1": "658645717510",
"us-west-2": "658645717510"
},
"repository": "smdistributed-modelparallel"
}
}
}
Expand Down
1 change: 1 addition & 0 deletions src/sagemaker/image_uris.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -699,6 +699,7 @@ def get_training_image_uri(
or "2.1" in framework_version
or "2.2" in framework_version
or "2.3" in framework_version
or "2.4" in framework_version
):
container_version = "cu121"
else:
Expand Down
11 changes: 7 additions & 4 deletions tests/unit/sagemaker/image_uris/test_smp_v2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,14 +36,17 @@ def test_smp_v2(load_config):
for region in ACCOUNTS.keys():
for instance_type in CONTAINER_VERSIONS.keys():
cuda_vers = CONTAINER_VERSIONS[instance_type]
if "2.1" in version or "2.2" in version or "2.3" in version:
if (
"2.1" in version
or "2.2" in version
or "2.3" in version
or "2.4" in version
):
cuda_vers = "cu121"

if "2.3.1" == version:
if "2.3.1" == version or "2.4.1" == version:
py_version = "py311"

print(version, py_version)

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