Re-enable HITS in the python API using the new primitive-based implem…

…entation (#1941)

Followup PR to #1930

This PR re-enables HITS in the python API using the new C++ primitive-based implementation. This also refactors the tests to use the benchmark fixture plugin, and adds an additional dataset to read for more diverse comparison to Nx.

![image](https://user-images.githubusercontent.com/3039903/141236018-16557063-9d2a-4fd7-b8c1-789f78958ea7.png)

Authors:
  - Rick Ratzel (https://github.com/rlratzel)

Approvers:
  - Chuck Hastings (https://github.com/ChuckHastings)
  - Brad Rees (https://github.com/BradReesWork)
  - Joseph Nke (https://github.com/jnke2016)

URL: #1941

cpp/include/cugraph/utilities/cython.hpp

@@ -524,6 +524,17 @@ void call_wcc(raft::handle_t const& handle,
  graph_container_t const& graph_container,
  vertex_t* components);
 
+// Wrapper for calling HITS through a graph container
+template <typename vertex_t, typename weight_t>
+void call_hits(raft::handle_t const& handle,
+ graph_container_t const& graph_container,
+ weight_t* hubs,
+ weight_t* authorities,
+ size_t max_iter,
+ weight_t tolerance,
+ const weight_t* starting_value,
+ bool normalized);
+
 // Wrapper for calling graph generator
 template <typename vertex_t>
 std::unique_ptr<graph_generator_t> call_generate_rmat_edgelist(raft::handle_t const& handle,

cpp/src/utilities/cython.cu

@@ -1098,6 +1098,84 @@ void call_wcc(raft::handle_t const& handle,
  }
 }
 
+// wrapper for HITS:
+//
+template <typename vertex_t, typename weight_t>
+void call_hits(raft::handle_t const& handle,
+ graph_container_t const& graph_container,
+ weight_t* hubs,
+ weight_t* authorities,
+ size_t max_iter,
+ weight_t tolerance,
+ const weight_t* starting_value,
+ bool normalized)
+{
+ constexpr bool has_initial_hubs_guess{false};
+ constexpr bool normalize{true};
+ constexpr bool do_expensive_check{false};
+ constexpr bool transposed{true};
+
+ // FIXME: most of these branches are not currently executed: MG support is not
+ // yet in the python API, and only int32_t edge types are being used. Consider
+ // removing these until actually needed.
+
+ if (graph_container.is_multi_gpu) {
+ constexpr bool multi_gpu{true};
+ if (graph_container.edgeType == numberTypeEnum::int32Type) {
+ auto graph = detail::create_graph<int32_t, int32_t, weight_t, transposed, multi_gpu>(
+ handle, graph_container);
+ cugraph::hits(handle,
+ graph->view(),
+ reinterpret_cast<weight_t*>(hubs),
+ reinterpret_cast<weight_t*>(authorities),
+ tolerance,
+ max_iter,
+ has_initial_hubs_guess,
+ normalize,
+ do_expensive_check);
+ } else if (graph_container.edgeType == numberTypeEnum::int64Type) {
+ auto graph = detail::create_graph<vertex_t, int64_t, weight_t, transposed, multi_gpu>(
+ handle, graph_container);
+ cugraph::hits(handle,
+ graph->view(),
+ reinterpret_cast<weight_t*>(hubs),
+ reinterpret_cast<weight_t*>(authorities),
+ tolerance,
+ max_iter,
+ has_initial_hubs_guess,
+ normalize,
+ do_expensive_check);
+ }
+ } else {
+ constexpr bool multi_gpu{false};
+ if (graph_container.edgeType == numberTypeEnum::int32Type) {
+ auto graph = detail::create_graph<int32_t, int32_t, weight_t, transposed, multi_gpu>(
+ handle, graph_container);
+ cugraph::hits(handle,
+ graph->view(),
+ reinterpret_cast<weight_t*>(hubs),
+ reinterpret_cast<weight_t*>(authorities),
+ tolerance,
+ max_iter,
+ has_initial_hubs_guess,
+ normalize,
+ do_expensive_check);
+ } else if (graph_container.edgeType == numberTypeEnum::int64Type) {
+ auto graph = detail::create_graph<vertex_t, int64_t, weight_t, transposed, multi_gpu>(
+ handle, graph_container);
+ cugraph::hits(handle,
+ graph->view(),
+ reinterpret_cast<weight_t*>(hubs),
+ reinterpret_cast<weight_t*>(authorities),
+ tolerance,
+ max_iter,
+ has_initial_hubs_guess,
+ normalize,
+ do_expensive_check);
+ }
+ }
+}
+
 // wrapper for shuffling:
 //
 template <typename vertex_t, typename edge_t, typename weight_t>
@@ -1509,6 +1587,42 @@ template void call_wcc<int64_t, double>(raft::handle_t const& handle,
  graph_container_t const& graph_container,
  int64_t* components);
 
+template void call_hits<int32_t, float>(raft::handle_t const& handle,
+ graph_container_t const& graph_container,
+ float* hubs,
+ float* authorities,
+ size_t max_iter,
+ float tolerance,
+ const float* starting_value,
+ bool normalized);
+
+template void call_hits<int32_t, double>(raft::handle_t const& handle,
+ graph_container_t const& graph_container,
+ double* hubs,
+ double* authorities,
+ size_t max_iter,
+ double tolerance,
+ const double* starting_value,
+ bool normalized);
+
+template void call_hits<int64_t, float>(raft::handle_t const& handle,
+ graph_container_t const& graph_container,
+ float* hubs,
+ float* authorities,
+ size_t max_iter,
+ float tolerance,
+ const float* starting_value,
+ bool normalized);
+
+template void call_hits<int64_t, double>(raft::handle_t const& handle,
+ graph_container_t const& graph_container,
+ double* hubs,
+ double* authorities,
+ size_t max_iter,
+ double tolerance,
+ const double* starting_value,
+ bool normalized);
+
 template std::unique_ptr<major_minor_weights_t<int32_t, int32_t, float>> call_shuffle(
  raft::handle_t const& handle,
  int32_t* edgelist_major_vertices,

notebooks/link_analysis/HITS.ipynb

@@ -5,7 +5,6 @@
  "metadata": {},
  "source": [
  "# HITS\n",
- "# Skip notebook test\n",
  "\n",
  "In this notebook, we will use both NetworkX and cuGraph to compute HITS. \n",
  "The NetworkX and cuGraph processes will be interleaved so that each step can be compared.\n",

python/cugraph/cugraph/centrality/betweenness_centrality.py

@@ -48,7 +48,7 @@ def betweenness_centrality(
  Parameters
  ----------
  G : cuGraph.Graph or networkx.Graph
- The graph can be either directed (DiGraph) or undirected (Graph).
+ The graph can be either directed (Graph(directed=True)) or undirected.
  Weights in the graph are ignored, the current implementation uses
  BFS traversals. Use weight parameter if weights need to be considered
  (currently not supported)
@@ -65,8 +65,8 @@ def betweenness_centrality(
  normalized : bool, optional
  Default is True.
  If true, the betweenness values are normalized by
- __2 / ((n - 1) * (n - 2))__ for Graphs (undirected), and
- __1 / ((n - 1) * (n - 2))__ for DiGraphs (directed graphs)
+ __2 / ((n - 1) * (n - 2))__ for undirected Graphs, and
+ __1 / ((n - 1) * (n - 2))__ for directed Graphs
  where n is the number of nodes in G.
  Normalization will ensure that values are in [0, 1],
  this normalization scales for the highest possible value where one
@@ -170,7 +170,7 @@ def edge_betweenness_centrality(
  Parameters
  ----------
  G : cuGraph.Graph or networkx.Graph
- The graph can be either directed (DiGraph) or undirected (Graph).
+ The graph can be either directed (Graph(directed=True)) or undirected.
  Weights in the graph are ignored, the current implementation uses
  BFS traversals. Use weight parameter if weights need to be considered
  (currently not supported)
@@ -186,8 +186,8 @@ def edge_betweenness_centrality(
  normalized : bool, optional
  Default is True.
  If true, the betweenness values are normalized by
- 2 / (n * (n - 1)) for Graphs (undirected), and
- 1 / (n * (n - 1)) for DiGraphs (directed graphs)
+ 2 / (n * (n - 1)) for undirected Graphs, and
+ 1 / (n * (n - 1)) for directed Graphs
  where n is the number of nodes in G.
  Normalization will ensure that values are in [0, 1],
  this normalization scales for the highest possible value where one

python/cugraph/cugraph/centrality/betweenness_centrality_wrapper.pyx

@@ -17,7 +17,6 @@
 # cython: language_level = 3
 
 from cugraph.centrality.betweenness_centrality cimport betweenness_centrality as c_betweenness_centrality
-from cugraph.structure.graph_classes import DiGraph
 from cugraph.structure.graph_primtypes cimport *
 from libc.stdint cimport uintptr_t
 from libcpp cimport bool
@@ -177,8 +176,7 @@ def batch_betweenness_centrality(input_graph, normalized, endpoints,
  comms = Comms.get_comms()
  replicated_adjlists = input_graph.batch_adjlists
  work_futures = [client.submit(run_mg_work,
- (data, type(input_graph)
- is DiGraph),
+ (data, input_graph.is_directed()),
  normalized,
  endpoints,
  weights,
@@ -197,7 +195,7 @@ def sg_betweenness_centrality(input_graph, normalized, endpoints, weights,
  handle = Comms.get_default_handle()
  adjlist = input_graph.adjlist
  input_data = ((adjlist.offsets, adjlist.indices, adjlist.weights),
- type(input_graph) is DiGraph)
+ input_graph.is_directed())
  df = run_internal_work(handle, input_data, normalized, endpoints, weights,
  vertices, result_dtype)
  return df

python/cugraph/cugraph/centrality/edge_betweenness_centrality_wrapper.pyx

@@ -18,7 +18,6 @@
 
 from cugraph.centrality.betweenness_centrality cimport edge_betweenness_centrality as c_edge_betweenness_centrality
 from cugraph.structure import graph_primtypes_wrapper
-from cugraph.structure.graph_classes import DiGraph, Graph
 from cugraph.structure.graph_primtypes cimport *
 from libc.stdint cimport uintptr_t
 from libcpp cimport bool
@@ -166,8 +165,7 @@ def batch_edge_betweenness_centrality(input_graph,
  comms = Comms.get_comms()
  replicated_adjlists = input_graph.batch_adjlists
  work_futures = [client.submit(run_mg_work,
- (data, type(input_graph)
- is DiGraph),
+ (data, input_graph.is_directed()),
  normalized,
  weights,
  vertices,
@@ -188,7 +186,7 @@ def sg_edge_betweenness_centrality(input_graph, normalized, weights,
  handle = Comms.get_default_handle()
  adjlist = input_graph.adjlist
  input_data = ((adjlist.offsets, adjlist.indices, adjlist.weights),
- type(input_graph) is DiGraph)
+ input_graph.is_directed())
  df = run_internal_work(handle, input_data, normalized, weights,
  vertices, result_dtype)
  return df

python/cugraph/cugraph/link_analysis/hits.pxd

@@ -11,22 +11,19 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-# cython: profile=False
-# distutils: language = c++
-# cython: embedsignature = True
-# cython: language_level = 3
-
-from cugraph.structure.graph_primtypes cimport *
 from libcpp cimport bool
 
+from cugraph.structure.graph_utilities cimport graph_container_t
+from cugraph.raft.common.handle cimport handle_t
 
-cdef extern from "cugraph/algorithms.hpp" namespace "cugraph::gunrock":
 
- cdef void hits[VT,ET,WT](
- const GraphCSRView[VT,ET,WT] &graph,
+cdef extern from "cugraph/utilities/cython.hpp" namespace "cugraph::cython":
+ cdef void call_hits[vertex_t,weight_t](
+ const handle_t &handle,
+ const graph_container_t &g,
+ weight_t *hubs,
+ weight_t *authorities,
  int max_iter,
- WT tolerance,
- const WT *starting_value,
- bool normalized,
- WT *hubs,
- WT *authorities) except +
+ weight_t tolerance,
+ const weight_t *starting_value,
+ bool normalized) except +

python/cugraph/cugraph/link_analysis/hits.py

@@ -11,11 +11,10 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-# from cugraph.link_analysis import hits_wrapper
-
-# from cugraph.utilities import (ensure_cugraph_obj_for_nx,
-# df_score_to_dictionary,
-# )
+from cugraph.link_analysis import hits_wrapper
+from cugraph.utilities import (ensure_cugraph_obj_for_nx,
+ df_score_to_dictionary,
+ )
 
 
 def hits(G, max_iter=100, tol=1.0e-5, nstart=None, normalized=True):
@@ -76,10 +75,9 @@ def hits(G, max_iter=100, tol=1.0e-5, nstart=None, normalized=True):
  >>> hits = cugraph.hits(G, max_iter = 50)
  """
 
- """
  G, isNx = ensure_cugraph_obj_for_nx(G)
 
- df = hits_wrapper.hits(G, max_iter, tol) # noqa: F821
+ df = hits_wrapper.hits(G, max_iter, tol)
 
  if G.renumbered:
  df = G.unrenumber(df, "vertex")
@@ -91,5 +89,3 @@ def hits(G, max_iter=100, tol=1.0e-5, nstart=None, normalized=True):
  df = (d1, d2)
 
  return df
- """
- raise NotImplementedError("Temporarily disabled. New version in 21.12")