You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have a custom distributed plugin, but it currently does not work PTL's automatic distributed sampler.
Plugin looks like this:
class MyPlugin(ParallelPlugin):
@property
def distributed_sampler_kwargs(self):
...
But in data_loading.py, when deciding whether to add a distributed data loader, PTL looks at accelerator_connector.is_distributed:
need_dist_sampler = self.accelerator_connector.is_distributed and not isinstance(
dataloader.sampler, DistributedSampler
)
And self.accelerator_connector.is_distributed only returns True if the built-in plugins are used, not any custom plugin:
@property
def is_distributed(self) -> bool:
is_distributed = self.use_ddp or self.use_ddp2 or self.use_horovod
if self.on_tpu:
is_distributed |= self.training_type_plugin.is_distributed
return is_distributed
Therefore, with a custom plugin, the distributed sampler is not set.
How can a custom plugin set itself to be distributed, so this property, and any other properties related to distributed training will automatically be set to the correct value?
🐛 Bug
I have a custom distributed plugin, but it currently does not work PTL's automatic distributed sampler.
Plugin looks like this:
But in
data_loading.py
, when deciding whether to add a distributed data loader, PTL looks ataccelerator_connector.is_distributed
:And
self.accelerator_connector.is_distributed
only returnsTrue
if the built-in plugins are used, not any custom plugin:Therefore, with a custom plugin, the distributed sampler is not set.
How can a custom plugin set itself to be distributed, so this property, and any other properties related to distributed training will automatically be set to the correct value?
cc @SeanNaren @justusschock @awaelchli
The text was updated successfully, but these errors were encountered: