Add logging for partitioner methods (#3290)

Summary:
Pull Request resolved: #3290

Adding logging to make it clear which partitioner method and its settings were called. Right now there is no way to tell that from the logs which occasionally confuses me...

Reviewed By: aliafzal

Differential Revision: D80361377

fbshipit-source-id: 91c6b5d27e433e66952ed949f113d34b0c8c9a4d

Author: Caner Gocmen

torchrec/distributed/planner/partitioners.py

@@ -226,6 +226,9 @@ def partition(
             # The topology updates are done after the end of all the placements (the other
             # in the example is just for clarity).
         """
+        logger.info(
+            f"GreedyPerfPartitioner - sort_by: {self._sort_by}, balance_modules: {self._balance_modules}"
+        )
 
         _topology: Topology = copy.deepcopy(storage_constraint)
         minheap_devices: Optional[List[OrderedDeviceHardware]] = None
@@ -587,6 +590,10 @@ def partition(
         within the tolerance of the original plan that uses the least amount
         of memory.
         """
+        logger.info(
+            f"MemoryBalancedPartitioner - _max_search_count: {self._max_search_count}, _tolerance: {self._tolerance}, _balance_modules: {self._balance_modules}"
+        )
+
         _perf_model: PerfModel = NoopPerfModel(storage_constraint)
         _partitioner = GreedyPerfPartitioner(
             sort_by=SortBy.PERF, balance_modules=self._balance_modules