Skip to content

Pin version of NGC Pytorch image to 25.06-py3 #386

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 3 commits into from
Sep 1, 2025
Merged

Pin version of NGC Pytorch image to 25.06-py3 #386

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
c6891aa
Pin version of NGC Pytorch image to 25.06-py3
Aug 28, 2025
20cf7c8
Make linter happy
Aug 28, 2025
8b3a7ab
Merge branch 'main' into pytorch_ngc_25.06
Sep 1, 2025
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
25 changes: 15 additions & 10 deletions checks/apps/pytorch/pytorch_allreduce.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
# SPDX-License-Identifier: BSD-3-Clause

import pathlib
import re
import re # noqa: F401
import sys

import reframe as rfm
Expand All @@ -26,22 +26,27 @@ class PyTorchNCCLAllReduce(rfm.RunOnlyRegressionTest, ContainerEngineMixin):
valid_prog_environs = ['builtin']
num_nodes = variable(int, value=8)
sourcesdir = None
curated_images = ['nvcr.io#nvidia/pytorch:24.12-py3']
curated_images = ['nvcr.io#nvidia/pytorch:25.06-py3']

# NOTE: only the "-py3" image is supported by the test
supported_flavors = ["-py3"]


pytorch_tags = nvidia_image_tags('pytorch')
latest_tags = []

for flavor in supported_flavors:
versions = []
for tag in pytorch_tags:
if re.match(rf'^\d+\.\d+{flavor}$', tag):
versions.append(tag[:-len(flavor)])
if versions:
latest_version = max(versions)
latest_tags += [f'{latest_version+flavor}']
# FIXME: 25.08-py3 version and above use Cuda 13 see:
# https://jira.cscs.ch/browse/VCUE-1039

# for flavor in supported_flavors:
# versions = []
# for tag in pytorch_tags:
# if re.match(rf'^\d+\.\d+{flavor}$', tag):
# versions.append(tag[:-len(flavor)])
# if versions:
# versions.sort(reverse=True)
# for v in versions:
# latest_tags += [f'{latest_version+flavor}']

latest_images = [f'nvcr.io#nvidia/pytorch:{tag}' for tag in latest_tags]
image = parameter(curated_images + latest_images)
Expand Down
23 changes: 13 additions & 10 deletions checks/apps/pytorch/pytorch_nvidia.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
import re
import re # noqa: F401
import sys
import pathlib
import reframe as rfm
Expand Down Expand Up @@ -38,23 +38,26 @@ class PyTorchDdpCeNv(PyTorchTestBase, ContainerEngineMixin):
'NVIDIA NGC')
valid_systems = ['+ce +nvgpu']
aws_ofi_nccl = parameter([True])
curated_images = ['nvcr.io#nvidia/pytorch:24.01-py3']
curated_images = ['nvcr.io#nvidia/pytorch:25.06-py3']

# NOTE: only the "-py3" image is supported by the test
supported_flavors = ["-py3"]

pytorch_tags = nvidia_image_tags('pytorch')
latest_tags = []

for flavor in supported_flavors:
versions = []
for tag in pytorch_tags:
if re.match(rf'^\d+\.\d+{flavor}$', tag):
versions.append(tag[:-len(flavor)])
# FIXME: 25.08-py3 version and above use Cuda 13 see:
# https://jira.cscs.ch/browse/VCUE-1039

if versions:
latest_version = max(versions)
latest_tags += [f'{latest_version}{flavor}']
# for flavor in supported_flavors:
# versions = []
# for tag in pytorch_tags:
# if re.match(rf'^\d+\.\d+{flavor}$', tag):
# versions.append(tag[:-len(flavor)])

# if versions:
# latest_version = max(versions)
# latest_tags += [f'{latest_version}{flavor}']

latest_images = [f'nvcr.io#nvidia/pytorch:{tag}' for tag in latest_tags]
image = parameter(curated_images + latest_images)
Expand Down