mathewjacob1002 commented on code in PR #41770:
URL: https://github.com/apache/spark/pull/41770#discussion_r1255035266
##########
python/pyspark/ml/torch/distributor.py:
##########
@@ -1003,3 +1007,97 @@ def _get_spark_partition_data_loader(
# if num_workers is zero, we cannot set `prefetch_factor` otherwise
# torch will raise error.
return DataLoader(dataset, batch_size, num_workers=num_workers)
+
+
+class DeepspeedTorchDistributor(TorchDistributor):
+
+ def __init__(self, num_processes: int = 1, local_mode: bool = True,
use_gpu: bool = True, deepspeed_config = None):
+ super().__init__(num_processes, local_mode, use_gpu)
+ self.deepspeed_config = deepspeed_config
+ self.ssl_conf = "deepspeed.spark.distributor.ignoreSsl"
Review Comment:
Assuming you're talking about the deepspeed_config: No we don't. It's if the
user already has a config file they would want to use, then they would pass
that in. Otherwise, they'd pass in a dictionary with arguments and we'd create
the conf file for them. Deepspeed will just query information if there isn't a
conf file passed in iirc.
Update: realized you're probably referencing the ssl_conf: it's put here to
differentiate from the TorchDistributor's ssl_conf
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
