DDPSpawnShardedPlugin¶
- class pytorch_lightning.plugins.training_type.DDPSpawnShardedPlugin(parallel_devices=None, num_nodes=None, cluster_environment=None, checkpoint_io=None, sync_batchnorm=None, ddp_comm_state=None, ddp_comm_hook=None, ddp_comm_wrapper=None, **kwargs)[source]¶
Bases:
pytorch_lightning.plugins.training_type.ddp_spawn.DDPSpawnPlugin
Optimizer sharded training provided by FairScale.
- block_backward_sync()[source]¶
Blocks syncing gradients behaviour on backwards pass.
This is useful for skipping sync when accumulating gradients, reducing communication overhead Returns: context manager with sync behaviour off
- Return type
- property lightning_module: pytorch_lightning.core.lightning.LightningModule¶
Returns the pure LightningModule without potential wrappers.