ClearML/clearml-agent
1
0
Fork 0
mirror of https://github.com/clearml/clearml-agent synced 2025-06-26 18:16:15 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Support NVIDIA_VISIBLE_DEVICES using volume mounts

Browse Source
This commit is contained in:
clearml 2025-06-22 22:37:27 +03:00
parent 768ee3d2cf
commit 553c72e06a
1 changed files with 10 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
clearml_agent/helper/resource_monitor.py
View File

@ -2,6 +2,7 @@ from __future__ import unicode_literals, division
import logging
import re
import os
import shlex
from collections import deque
from itertools import starmap
@ -112,7 +113,15 @@ class ResourceMonitor(object):
active_gpus = Session.get_nvidia_visible_env()
# None means no filtering, report all gpus
if active_gpus and active_gpus != "all":
self._active_gpus = [g.strip() for g in str(active_gpus).split(',')]
if os.path.isdir(active_gpus):
try:
self._active_gpus = os.listdir(active_gpus)
except OSError as e:
log.warning(
"Failed listing {}: {}".format(active_gpus, e)
)
else:
self._active_gpus = [g.strip() for g in active_gpus.split(",")]
except Exception:
pass
self._cluster_report_interval_sec = int(session.config.get(