# Functions
ContainsMasterSpec returns true if the pytorchjob contains master spec.
No description provided by the author
No description provided by the author
NewReconciler creates a PyTorchJob Reconciler.
# Constants
EnvMaxRestarts is the environment variable name for the maximum number of worker group restarts before failing.
EnvMonitorInterval is the environment variable name for the interval, in seconds, to monitor the state of workers.
EnvNnodes is the environment variable name for the number of nodes.
EnvNodeRank is the environment variable name for the rank of nodes.
EnvNprocPerNode is the environment variable name for the number of processes per node.
EnvNProcPerNode is the environment variable name for the number of processes per node.
EnvRDZVBackend is the environment variable name for the rdzv backend.
ENVRDZVConf is the environment variable name for the rdzv conf.
EnvRDZVEndpoint is the environment variable name for the rdzv endpoint.
EnvRDZVID is the environment variable name for the rdzv id.
EnvRDZVStandalone is the environment variable name for the standalone mode.
EnvStartMethod is the environment variable name for the multiprocessing start method to use when creating workers, which could be fork, spawn and forkserver.
# Variables
No description provided by the author
No description provided by the author
No description provided by the author
No description provided by the author
# Structs
ElasticEnvVarGenerator is the environment variable generator for Elastic related arguments.
MasterEnvVarGenerator is the environment variable generator for Master related arguments.
PyTorchJobReconciler reconciles a PyTorchJob object.
# Interfaces
EnvVarGenerator is the environment variable generator interface.