rithwik-db commented on code in PR #39188:
URL: https://github.com/apache/spark/pull/39188#discussion_r1068924576
##########
python/pyspark/ml/torch/distributor.py:
##########
@@ -261,6 +312,130 @@ def __init__(
super().__init__(num_processes, local_mode, use_gpu)
self.ssl_conf = "pytorch.spark.distributor.ignoreSsl" # type: ignore
self._validate_input_params()
+ self.input_params = self._create_input_params()
+
+ @staticmethod
+ def _create_torchrun_command(
+ input_params: dict[str, Any], path_to_train_file: str, *args: Any
+ ) -> list[str]:
+ local_mode = input_params["local_mode"]
+ num_processes = input_params["num_processes"]
+
+ if local_mode:
+ torchrun_args = ["--standalone", "--nnodes=1"]
+ processes_per_node = num_processes
+ else:
+ pass
+ # TODO(SPARK-41592): Handle distributed training
Review Comment:
`local_mode` will always be true for this PR, but the follow up
[PR](https://github.com/apache/spark/pull/39267) addresses the `else` section.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
