replicate latest pytorch and tensorflow images with gpu support
3 unresolved threads
3 unresolved threads
replicate latest pytorch and tensorflow images with gpu support; fix https://gitlab.cern.ch/mlops/project/-/issues/144
Merge request reports
Activity
Filter activity
requested review from @rchiores
310 310 gcr.io/knative-releases/knative.dev/serving/cmd/webhook@sha256:48aee2733721ecc77956abc5a2ca072853a669ebc97519beb48f7b3da8455e67 311 311 gcr.io/ml-pipeline/workflow-controller:v3.4.17-license-compliance 312 312 docker.io/kserve/xgbserver:v0.13.1 313 docker.io/pytorch/pytorch:2.6.0-cuda12.6-cudnn9-runtime changed this line in version 2 of the diff
we are currently using
CUDA Version: 12.4
on all 6 gpu nodes, I doubt we would want any image for a version older than that?Edited by Raulian-Ionut Chiorescu294 294 docker.io/kubeflowkatib/suggestion-optuna:v0.17.0 295 295 docker.io/kubeflowkatib/suggestion-pbt:v0.17.0 296 296 docker.io/kubeflowkatib/suggestion-skopt:v0.17.0 297 297 docker.io/kubeflownotebookswg/tensorboard-controller:v1.9.2 298 298 docker.io/kubeflownotebookswg/tensorboards-web-app:v1.9.2
Please register or sign in to reply