Python code cleanup across docs, wrappers, testing (#2194)

docs/cugraph/source/api_docs/centrality.rst

@@ -26,4 +26,3 @@ Katz Centrality (MG)
    :toctree: api/
 
    cugraph.dask.centrality.katz_centrality.katz_centrality
-   cugraph.dask.centrality.katz_centrality.call_katz_centrality

docs/cugraph/source/api_docs/community.rst

@@ -50,7 +50,6 @@ Louvain (MG)
 .. autosummary::
    :toctree: api/
 
-   cugraph.dask.community.louvain.call_louvain
    cugraph.dask.community.louvain.louvain
 
 Spectral Clustering

docs/cugraph/source/api_docs/components.rst

@@ -20,6 +20,5 @@ Connected Components (MG)
 .. autosummary::
    :toctree: api/
 
-   cugraph.dask.components.connectivity.call_wcc
    cugraph.dask.components.connectivity.weakly_connected_components
 

docs/cugraph/source/api_docs/helper_functions.rst

@@ -12,5 +12,13 @@ Methods
 
    cugraph.comms.comms.initialize
    cugraph.comms.comms.destroy
+   cugraph.comms.comms.is_initialized
+   cugraph.comms.comms.get_comms
+   cugraph.comms.comms.get_workers
+   cugraph.comms.comms.get_session_id
+   cugraph.comms.comms.get_2D_partition
+   cugraph.comms.comms.get_default_handle
+   cugraph.comms.comms.get_handle
+   cugraph.comms.comms.get_worker_id
    cugraph.dask.common.read_utils.get_chunksize
 

docs/cugraph/source/api_docs/link_analysis.rst

@@ -12,6 +12,13 @@ HITS
 
    cugraph.hits
 
+HITS (MG)
+---------
+.. autosummary::
+   :toctree: api/
+
+   cugraph.dask.link_analysis.hits.hits
+
 
 Pagerank
 --------

docs/cugraph/source/api_docs/sampling.rst

@@ -12,4 +12,11 @@ Random Walks
 
    cugraph.random_walks
    cugraph.ego_graph
-   cugraph.experimental.dask.uniform_neighborhood_sampling
+   cugraph.experimental.dask.uniform_neighborhood_sampling
+
+Node2Vec
+---------
+.. autosummary::
+   :toctree: api/
+
+   cugraph.node2vec

docs/cugraph/source/api_docs/structure.rst

@@ -75,11 +75,30 @@ Conversion from Other Formats
    cugraph.to_pandas_adjacency
    cugraph.to_pandas_edgelist
 
+NumberMap
+-----------------------------
+.. autosummary::
+   :toctree: api/
+
+   cugraph.structure.NumberMap
+   cugraph.structure.NumberMap.MultiGPU
+   cugraph.structure.NumberMap.SingleGPU
+   cugraph.structure.NumberMap.from_internal_vertex_id
+   cugraph.structure.NumberMap.to_internal_vertex_id
+   cugraph.structure.NumberMap.add_internal_vertex_id
+   cugraph.structure.NumberMap.compute_vals
+   cugraph.structure.NumberMap.compute_vals_types
+   cugraph.structure.NumberMap.generate_unused_column_name
+   cugraph.structure.NumberMap.renumber
+   cugraph.structure.NumberMap.renumber_and_segment
+   cugraph.structure.NumberMap.set_renumbered_col_names
+   cugraph.structure.NumberMap.unrenumber
+   cugraph.structure.NumberMap.vertex_column_size
+
 Other
 -----------------------------
 .. autosummary::
    :toctree: api/
 
    cugraph.hypergraph
    cugraph.structure.shuffle
-   cugraph.structure.NumberMap

docs/cugraph/source/api_docs/traversal.rst

@@ -19,7 +19,6 @@ Breadth-first-search (MG)
    :toctree: api/
 
    cugraph.dask.traversal.bfs.bfs
-   cugraph.dask.traversal.bfs.call_bfs
 
 Single-source-shortest-path
 ---------------------------
@@ -36,5 +35,4 @@ Single-source-shortest-path (MG)
 .. autosummary::
    :toctree: api/
 
-   cugraph.dask.traversal.sssp.call_sssp
    cugraph.dask.traversal.sssp.sssp

python/cugraph/cugraph/centrality/betweenness_centrality.py

@@ -81,7 +81,7 @@ def betweenness_centrality(
         If true, include the endpoints in the shortest path counts.
         (Not Supported)
 
-    seed : optional
+    seed : optional (default=None)
         if k is specified and k is an integer, use seed to initialize the
         random number generator.
         Using None as seed relies on random.seed() behavior: using current

python/cugraph/cugraph/comms/comms.py

@@ -108,6 +108,21 @@ def initialize(comms=None,
         default) represents a partitioning resulting in prows*pcols
         partitions. A non-1 value currently results in a partitioning of
         p*pcols partitions, where p is the number of GPUs.
+
+    Examples
+    --------
+    >>> from dask.distributed import Client
+    >>> from dask_cuda import LocalCUDACluster
+    >>> import cugraph.comms as Comms
+    >>> cluster = LocalCUDACluster()
+    >>> client = Client(cluster)
+    >>> Comms.initialize(p2p=True)
+    >>> # DO WORK HERE
+    >>> # All done, clean up
+    >>> Comms.destroy()
+    >>> client.close()
+    >>> cluster.close()
+
     """
 
     global __instance
@@ -200,11 +215,17 @@ def get_default_handle():
 # Functions to be called from within workers
 
 def get_handle(sID):
+    """
+    Returns the handle from within the worker using the sessionstate.
+    """
     sessionstate = get_raft_comm_state(sID)
     return sessionstate['handle']
 
 
 def get_worker_id(sID):
+    """
+    Returns the worker's sessionId from within the worker.
+    """
     sessionstate = get_raft_comm_state(sID)
     return sessionstate['wid']
 

python/cugraph/cugraph/community/egonet.py

@@ -163,7 +163,8 @@ def batched_ego_graphs(
     ...                   header=None)
     >>> G = cugraph.Graph()
     >>> G.from_cudf_edgelist(M, source='0', destination='1')
-    >>> b_ego_graph = cugraph.batched_ego_graphs(G, seeds=[1,5], radius=2)
+    >>> b_ego_graph, offsets = cugraph.batched_ego_graphs(G, seeds=[1,5],
+    ...                                                   radius=2)
 
     """
 

python/cugraph/cugraph/community/ktruss_subgraph.py

@@ -64,6 +64,15 @@ def k_truss(G, k):
     G_truss : cuGraph.Graph or networkx.Graph
         A cugraph graph descriptor with the k-truss subgraph for the given k.
         The networkx graph will NOT have all attributes copied over
+
+    Examples
+    --------
+    >>> gdf = cudf.read_csv(datasets_path / 'karate.csv', delimiter=' ',
+    ...                     dtype=['int32', 'int32', 'float32'], header=None)
+    >>> G = cugraph.Graph()
+    >>> G.from_cudf_edgelist(gdf, source='0', destination='1')
+    >>> k_subgraph = cugraph.k_truss(G, 3)
+
     """
 
     _ensure_compatible_cuda_version()

python/cugraph/cugraph/dask/centrality/katz_centrality.py

@@ -132,20 +132,21 @@ def katz_centrality(input_graph,
 
     Examples
     --------
-    >>> # import cugraph.dask as dcg
+    >>> import cugraph.dask as dcg
+    >>> import dask_cudf
     >>> # ... Init a DASK Cluster
     >>> #    see https://docs.rapids.ai/api/cugraph/stable/dask-cugraph.html
     >>> # Download dataset from https://github.com/rapidsai/cugraph/datasets/..
-    >>> # chunksize = dcg.get_chunksize(datasets_path / "karate.csv")
-    >>> # ddf = dask_cudf.read_csv(input_data_path, chunksize=chunksize)
-    >>> # dg = cugraph.Graph(directed=True)
-    >>> # dg.from_dask_cudf_edgelist(ddf, source='src', destination='dst',
-    >>> #                            edge_attr='value')
-    >>> # pr = dcg.katz_centrality(dg)
+    >>> chunksize = dcg.get_chunksize(datasets_path / "karate.csv")
+    >>> ddf = dask_cudf.read_csv(datasets_path / "karate.csv",
+    ...                          chunksize=chunksize, delimiter=" ",
+    ...                          names=["src", "dst", "value"],
+    ...                          dtype=["int32", "int32", "float32"])
+    >>> dg = cugraph.Graph(directed=True)
+    >>> dg.from_dask_cudf_edgelist(ddf, source='src', destination='dst')
+    >>> pr = dcg.katz_centrality(dg)
 
     """
-    # FIXME: Uncomment out the above (broken) example
-
     nstart = None
 
     client = default_client()

python/cugraph/cugraph/dask/common/read_utils.py

@@ -27,10 +27,9 @@ def get_chunksize(input_path):
     Examples
     --------
     >>> import cugraph.dask as dcg
-    >>> # chunksize = dcg.get_chunksize(edge_list.csv)
+    >>> chunksize = dcg.get_chunksize(datasets_path / 'netscience.csv')
 
     """
-    # FIXME: Uncomment out the above example
 
     import os
     from glob import glob

python/cugraph/cugraph/dask/community/louvain.py

@@ -19,7 +19,6 @@
 from cugraph.dask.common.input_utils import (get_distributed_data,
                                              get_vertex_partition_offsets)
 from cugraph.dask.community import louvain_wrapper as c_mg_louvain
-from cugraph.utilities.utils import is_cuda_version_less_than
 
 import dask_cudf
 
@@ -99,26 +98,22 @@ def louvain(input_graph, max_iter=100, resolution=1.0):
 
     Examples
     --------
-    >>> # import cugraph.dask as dcg
+    >>> import cugraph.dask as dcg
+    >>> import dask_cudf
     >>> # ... Init a DASK Cluster
     >>> #    see https://docs.rapids.ai/api/cugraph/stable/dask-cugraph.html
     >>> # Download dataset from https://github.com/rapidsai/cugraph/datasets/..
-    >>> # chunksize = dcg.get_chunksize(datasets_path / "karate.csv")
-    >>> # ddf = dask_cudf.read_csv(input_data_path, chunksize=chunksize)
-    >>> # dg = cugraph.Graph(directed=True)
-    >>> # dg.from_dask_cudf_edgelist(ddf, source='src', destination='dst',
-    >>> #                            edge_attr='value')
-    >>> # parts, modularity_score = dcg.louvain(dg)
-    """
-    # FIXME: Uncomment out the above (broken) example
-
-    # MG Louvain currently requires CUDA 10.2 or higher.
-    # FIXME: remove this check once RAPIDS drops support for CUDA < 10.2
-    if is_cuda_version_less_than((10, 2)):
-        raise NotImplementedError("Multi-GPU Louvain is not implemented for "
-                                  "this version of CUDA. Ensure CUDA version "
-                                  "10.2 or higher is installed.")
+    >>> chunksize = dcg.get_chunksize(datasets_path / "karate.csv")
+    >>> ddf = dask_cudf.read_csv(datasets_path / "karate.csv",
+    ...                          chunksize=chunksize, delimiter=" ",
+    ...                          names=["src", "dst", "value"],
+    ...                          dtype=["int32", "int32", "float32"])
+    >>> dg = cugraph.Graph(directed=True)
+    >>> dg.from_dask_cudf_edgelist(ddf, source='src', destination='dst',
+    ...                            edge_attr='value')
+    >>> parts, modularity_score = dcg.louvain(dg)
 
+    """
     # FIXME: dask methods to populate graphs from edgelists are only present on
     # DiGraph classes. Disable the Graph check for now and assume inputs are
     # symmetric DiGraphs.

python/cugraph/cugraph/dask/components/connectivity.py

@@ -54,6 +54,24 @@ def weakly_connected_components(input_graph):
 
         Graph or matrix object, which should contain the connectivity
         information
+
+    Examples
+    --------
+    >>> import cugraph.dask as dcg
+    >>> import dask_cudf
+    >>> # ... Init a DASK Cluster
+    >>> #    see https://docs.rapids.ai/api/cugraph/stable/dask-cugraph.html
+    >>> # Download dataset from https://github.com/rapidsai/cugraph/datasets/..
+    >>> chunksize = dcg.get_chunksize(datasets_path / "karate.csv")
+    >>> ddf = dask_cudf.read_csv(datasets_path / "karate.csv",
+    ...                          chunksize=chunksize, delimiter=" ",
+    ...                          names=["src", "dst", "value"],
+    ...                          dtype=["int32", "int32", "float32"])
+    >>> dg = cugraph.Graph(directed=True)
+    >>> dg.from_dask_cudf_edgelist(ddf, source='src', destination='dst',
+    ...                            edge_attr='value')
+    >>> result = dcg.weakly_connected_components(dg)
+
     """
 
     client = default_client()

python/cugraph/cugraph/dask/link_analysis/hits.py

@@ -135,16 +135,20 @@ def hits(input_graph, tol=1.0e-5, max_iter=100,  nstart=None, normalized=True):
 
     Examples
     --------
-    >>> # import cugraph.dask as dcg
+    >>> import cugraph.dask as dcg
+    >>> import dask_cudf
     >>> # ... Init a DASK Cluster
     >>> #    see https://docs.rapids.ai/api/cugraph/stable/dask-cugraph.html
     >>> # Download dataset from https://github.com/rapidsai/cugraph/datasets/..
-    >>> # chunksize = dcg.get_chunksize(datasets_path / "karate.csv")
-    >>> # ddf = dask_cudf.read_csv(input_data_path, chunksize=chunksize)
-    >>> # dg = cugraph.Graph(directed=True)
-    >>> # dg.from_dask_cudf_edgelist(ddf, source='src', destination='dst',
-    >>> #                            edge_attr='value')
-    >>> # hits = dcg.hits(dg, max_iter = 50)
+    >>> chunksize = dcg.get_chunksize(datasets_path / "karate.csv")
+    >>> ddf = dask_cudf.read_csv(datasets_path / "karate.csv",
+    ...                          chunksize=chunksize, delimiter=" ",
+    ...                          names=["src", "dst", "value"],
+    ...                          dtype=["int32", "int32", "float32"])
+    >>> dg = cugraph.Graph(directed=True)
+    >>> dg.from_dask_cudf_edgelist(ddf, source='src', destination='dst',
+    ...                            edge_attr='value')
+    >>> hits = dcg.hits(dg, max_iter = 50)
 
     """
 

python/cugraph/cugraph/dask/link_analysis/pagerank.py

@@ -121,16 +121,19 @@ def pagerank(input_graph,
 
     Examples
     --------
-    >>> # import cugraph.dask as dcg
+    >>> import cugraph.dask as dcg
+    >>> import dask_cudf
     >>> # ... Init a DASK Cluster
     >>> #    see https://docs.rapids.ai/api/cugraph/stable/dask-cugraph.html
     >>> # Download dataset from https://github.com/rapidsai/cugraph/datasets/..
-    >>> # chunksize = dcg.get_chunksize(datasets_path / "karate.csv")
-    >>> # ddf = dask_cudf.read_csv(input_data_path, chunksize=chunksize)
-    >>> # dg = cugraph.Graph(directed=True)
-    >>> # dg.from_dask_cudf_edgelist(ddf, source='src', destination='dst',
-    >>> #                            edge_attr='value')
-    >>> # pr = dcg.pagerank(dg)
+    >>> chunksize = dcg.get_chunksize(datasets_path / "karate.csv")
+    >>> ddf = dask_cudf.read_csv(datasets_path / "karate.csv",
+    ...                          chunksize=chunksize, delimiter=" ",
+    ...                          names=["src", "dst", "value"],
+    ...                          dtype=["int32", "int32", "float32"])
+    >>> dg = cugraph.Graph(directed=True)
+    >>> dg.from_dask_cudf_edgelist(ddf, source='src', destination='dst')
+    >>> pr = dcg.pagerank(dg)
 
     """
     nstart = None

python/cugraph/cugraph/dask/traversal/bfs.py

@@ -94,19 +94,22 @@ def bfs(input_graph,
 
     Examples
     --------
-    >>> # import cugraph.dask as dcg
+    >>> import cugraph.dask as dcg
+    >>> import dask_cudf
     >>> # ... Init a DASK Cluster
     >>> #    see https://docs.rapids.ai/api/cugraph/stable/dask-cugraph.html
     >>> # Download dataset from https://github.com/rapidsai/cugraph/datasets/..
-    >>> # chunksize = dcg.get_chunksize(datasets_path / "karate.csv")
-    >>> # ddf = dask_cudf.read_csv(input_data_path, chunksize=chunksize)
-    >>> # dg = cugraph.Graph(directed=True)
-    >>> # dg.from_dask_cudf_edgelist(ddf, source='src', destination='dst',
-    >>> #                            edge_attr='value')
-    >>> # df = dcg.bfs(dg, 0)
+    >>> chunksize = dcg.get_chunksize(datasets_path / "karate.csv")
+    >>> ddf = dask_cudf.read_csv(datasets_path / "karate.csv",
+    ...                          chunksize=chunksize, delimiter=" ",
+    ...                          names=["src", "dst", "value"],
+    ...                          dtype=["int32", "int32", "float32"])
+    >>> dg = cugraph.Graph(directed=True)
+    >>> dg.from_dask_cudf_edgelist(ddf, source='src', destination='dst',
+    ...                            edge_attr='value')
+    >>> df = dcg.bfs(dg, 0)
 
     """
-    # FIXME: Uncomment out the above (broken) example
 
     client = default_client()