Merge pull request #3783 from rapidsai/branch-23.08

Forward-merge branch-23.08 to branch-23.10

python/cugraph/cugraph/structure/graph_implementation/simpleDistributedGraph.py

@@ -181,6 +181,8 @@ def __from_edgelist(
         workers = _client.scheduler_info()["workers"]
         # Repartition to 2 partitions per GPU for memory efficient process
         input_ddf = input_ddf.repartition(npartitions=len(workers) * 2)
+        # FIXME: Make a copy of the input ddf before implicitly altering it.
+        input_ddf = input_ddf.map_partitions(lambda df: df.copy())
         # The dataframe will be symmetrized iff the graph is undirected
         # otherwise, the inital dataframe will be returned
         if edge_attr is not None:
@@ -317,7 +319,6 @@ def __from_edgelist(
             is_symmetric=not self.properties.directed,
         )
         ddf = ddf.repartition(npartitions=len(workers) * 2)
-        ddf = ddf.map_partitions(lambda df: df.copy())
         ddf = persist_dask_df_equal_parts_per_worker(ddf, _client)
         num_edges = len(ddf)
         ddf = get_persisted_df_worker_map(ddf, _client)