run_k8s_job

Function run_k8s_job 

Source
pub async fn run_k8s_job(
    job_id: &str,
    script: &str,
    image: &str,
    gpus_per_node: u32,
    nodes: u32,
    namespace: &str,
    args: &[String],
    log_dir: &Path,
) -> Result<i32>
Expand description

Submit a distributed training job to Kubernetes via PyTorchJob CRD.