How to use the torchelastic.train function in torchelastic
To help you get started, we’ve selected a few torchelastic examples, based on popular ways it is used in public projects.
Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.
pytorch / elastic / examples / imagenet / main.py View on Github 
coordinator = CoordinatorP2P(
c10d_backend=c10d_backend,
init_method=rdzv_init_url,
max_num_trainers=max_world_size,
process_group_timeout=60000,
)
state = ImagenetState(
model=model,
params=training_params,
dataset=train_dataset,
num_epochs=training_params.num_epochs,
)
log.info(f"Entering torchelastic train_loop")
torchelastic.train(coordinator, train_step, state)
Popular torchelastic functions
- torchelastic.coordinator.NonRetryableException
- torchelastic.coordinator.StopException
- torchelastic.distributed
- torchelastic.metrics
- torchelastic.metrics.get_elapsed_time_ms
- torchelastic.metrics.initialize_metrics
- torchelastic.metrics.profile
- torchelastic.metrics.publish_metric
- torchelastic.p2p.coordinator_p2p.CoordinatorP2P
- torchelastic.rendezvous.etcd_rendezvous.EtcdRendezvousRetryableFailure
- torchelastic.rendezvous.etcd_rendezvous.EtcdRendezvousRetryImmediately
- torchelastic.rendezvous.RendezvousClosedException
- torchelastic.rendezvous.RendezvousNonRetryableError
- torchelastic.rendezvous.RendezvousTimeoutException
- torchelastic.SimpleWorkerStats
- torchelastic.State
- torchelastic.train
- torchelastic.train_loop