Kubeflow Trainer torch runtime auto-injects torchrun only when command is not set in the container spec. Training script is now specified at TrainJob level via trainer.command instead. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> |
||
|---|---|---|
| .. | ||
| kubeflow-trainer-values.yaml | ||
| volcano-trainjob-integration.yaml | ||
| volcano-values.yaml | ||