Update API for node2vec and biased random walks (#4841)

This PR 1. Add MG support for node2vec random walks on the C, PLC and Python API 2. Fix renumbering bug in the CAPI for node2vec and biased random walks 3. Enable and fix bug in the MG C tests 4. Deprecate legacy node2vec and uniform random walks 5. Add SG and MG support for Biased random walks 6. Add multi GPU support for ECG in the python API closes #3277 closes #2558 closes #2559 closes #2560 closes #2561 closes #2853 closes #4843 Authors: - Joseph Nke (https://github.com/jnke2016) Approvers: - Chuck Hastings (https://github.com/ChuckHastings) - Rick Ratzel (https://github.com/rlratzel) URL: #4841
rapidsai · Feb 5, 2025 · dfdc73a · dfdc73a
1 parent 9d8fbd6
commit dfdc73a
Show file tree

Hide file tree

Showing 39 changed files with 3,416 additions and 113 deletions.
diff --git a/cpp/include/cugraph_c/sampling_algorithms.h b/cpp/include/cugraph_c/sampling_algorithms.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021-2024, NVIDIA CORPORATION.
+ * Copyright (c) 2021-2025, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -41,6 +41,7 @@ typedef struct {
  * @brief  Compute uniform random walks
  *
  * @param [in]  handle          Handle for accessing resources
+ * @param [in,out] rng_state    State of the random number generator, updated with each call
  * @param [in]  graph           Pointer to graph.  NOTE: Graph might be modified if the storage
  *                              needs to be transposed
  * @param [in]  start_vertices  Array of source vertices
@@ -52,6 +53,7 @@ typedef struct {
  */
 cugraph_error_code_t cugraph_uniform_random_walks(
   const cugraph_resource_handle_t* handle,
+  cugraph_rng_state_t* rng_state,
   cugraph_graph_t* graph,
   const cugraph_type_erased_device_array_view_t* start_vertices,
   size_t max_length,
@@ -62,6 +64,7 @@ cugraph_error_code_t cugraph_uniform_random_walks(
  * @brief  Compute biased random walks
  *
  * @param [in]  handle          Handle for accessing resources
+ * @param [in,out] rng_state    State of the random number generator, updated with each call
  * @param [in]  graph           Pointer to graph.  NOTE: Graph might be modified if the storage
  *                              needs to be transposed
  * @param [in]  start_vertices  Array of source vertices
@@ -73,6 +76,7 @@ cugraph_error_code_t cugraph_uniform_random_walks(
  */
 cugraph_error_code_t cugraph_biased_random_walks(
   const cugraph_resource_handle_t* handle,
+  cugraph_rng_state_t* rng_state,
   cugraph_graph_t* graph,
   const cugraph_type_erased_device_array_view_t* start_vertices,
   size_t max_length,
@@ -83,6 +87,7 @@ cugraph_error_code_t cugraph_biased_random_walks(
  * @brief  Compute random walks using the node2vec framework.
  *
  * @param [in]  handle          Handle for accessing resources
+ * @param [in,out] rng_state    State of the random number generator, updated with each call
  * @param [in]  graph           Pointer to graph.  NOTE: Graph might be modified if the storage
  *                              needs to be transposed
  * @param [in]  start_vertices  Array of source vertices
@@ -98,6 +103,7 @@ cugraph_error_code_t cugraph_biased_random_walks(
  */
 cugraph_error_code_t cugraph_node2vec_random_walks(
   const cugraph_resource_handle_t* handle,
+  cugraph_rng_state_t* rng_state,
   cugraph_graph_t* graph,
   const cugraph_type_erased_device_array_view_t* start_vertices,
   size_t max_length,

diff --git a/cpp/src/c_api/random_walks.cpp b/cpp/src/c_api/random_walks.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2022-2024, NVIDIA CORPORATION.
+ * Copyright (c) 2022-2025, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -154,19 +154,20 @@ namespace {
 
 struct uniform_random_walks_functor : public cugraph::c_api::abstract_functor {
   raft::handle_t const& handle_;
-  //  FIXME: rng_state_ should be passed as a parameter
   cugraph::c_api::cugraph_rng_state_t* rng_state_{nullptr};
   cugraph::c_api::cugraph_graph_t* graph_{nullptr};
   cugraph::c_api::cugraph_type_erased_device_array_view_t const* start_vertices_{nullptr};
   size_t max_length_{0};
   cugraph::c_api::cugraph_random_walk_result_t* result_{nullptr};
 
   uniform_random_walks_functor(cugraph_resource_handle_t const* handle,
+                               cugraph_rng_state_t* rng_state,
                                cugraph_graph_t* graph,
                                cugraph_type_erased_device_array_view_t const* start_vertices,
                                size_t max_length)
     : abstract_functor(),
       handle_(*reinterpret_cast<cugraph::c_api::cugraph_resource_handle_t const*>(handle)->handle_),
+      rng_state_(reinterpret_cast<cugraph::c_api::cugraph_rng_state_t*>(rng_state)),
       graph_(reinterpret_cast<cugraph::c_api::cugraph_graph_t*>(graph)),
       start_vertices_(
         reinterpret_cast<cugraph::c_api::cugraph_type_erased_device_array_view_t const*>(
@@ -224,10 +225,6 @@ struct uniform_random_walks_functor : public cugraph::c_api::abstract_functor {
         graph_view.local_vertex_partition_range_last(),
         false);
 
-      //  FIXME: remove once rng_state passed as parameter
-      rng_state_ = reinterpret_cast<cugraph::c_api::cugraph_rng_state_t*>(
-        new cugraph::c_api::cugraph_rng_state_t{raft::random::RngState{0}});
-
       auto [paths, weights] = cugraph::uniform_random_walks(
         handle_,
         rng_state_->rng_state_,
@@ -261,19 +258,20 @@ struct uniform_random_walks_functor : public cugraph::c_api::abstract_functor {
 
 struct biased_random_walks_functor : public cugraph::c_api::abstract_functor {
   raft::handle_t const& handle_;
-  //  FIXME: rng_state_ should be passed as a parameter
   cugraph::c_api::cugraph_rng_state_t* rng_state_{nullptr};
   cugraph::c_api::cugraph_graph_t* graph_{nullptr};
   cugraph::c_api::cugraph_type_erased_device_array_view_t const* start_vertices_{nullptr};
   size_t max_length_{0};
   cugraph::c_api::cugraph_random_walk_result_t* result_{nullptr};
 
   biased_random_walks_functor(cugraph_resource_handle_t const* handle,
+                              cugraph_rng_state_t* rng_state,
                               cugraph_graph_t* graph,
                               cugraph_type_erased_device_array_view_t const* start_vertices,
                               size_t max_length)
     : abstract_functor(),
       handle_(*reinterpret_cast<cugraph::c_api::cugraph_resource_handle_t const*>(handle)->handle_),
+      rng_state_(reinterpret_cast<cugraph::c_api::cugraph_rng_state_t*>(rng_state)),
       graph_(reinterpret_cast<cugraph::c_api::cugraph_graph_t*>(graph)),
       start_vertices_(
         reinterpret_cast<cugraph::c_api::cugraph_type_erased_device_array_view_t const*>(
@@ -293,8 +291,6 @@ struct biased_random_walks_functor : public cugraph::c_api::abstract_functor {
     // FIXME: Think about how to handle SG vice MG
     if constexpr (!cugraph::is_candidate<vertex_t, edge_t, weight_t>::value) {
       unsupported();
-    } else if constexpr (multi_gpu) {
-      unsupported();
     } else {
       // random walks expects store_transposed == false
       if constexpr (store_transposed) {
@@ -333,10 +329,6 @@ struct biased_random_walks_functor : public cugraph::c_api::abstract_functor {
         graph_view.local_vertex_partition_range_last(),
         false);
 
-      //  FIXME: remove once rng_state passed as parameter
-      rng_state_ = reinterpret_cast<cugraph::c_api::cugraph_rng_state_t*>(
-        new cugraph::c_api::cugraph_rng_state_t{raft::random::RngState{0}});
-
       auto [paths, weights] = cugraph::biased_random_walks(
         handle_,
         rng_state_->rng_state_,
@@ -348,8 +340,13 @@ struct biased_random_walks_functor : public cugraph::c_api::abstract_functor {
       //
       // Need to unrenumber the vertices in the resulting paths
       //
-      cugraph::unrenumber_local_int_vertices<vertex_t>(
-        handle_, paths.data(), paths.size(), number_map->data(), 0, paths.size() - 1, false);
+      cugraph::unrenumber_int_vertices<vertex_t, multi_gpu>(
+        handle_,
+        paths.data(),
+        paths.size(),
+        number_map->data(),
+        graph_view.vertex_partition_range_lasts(),
+        false);
 
       result_ = new cugraph::c_api::cugraph_random_walk_result_t{
         false,
@@ -365,7 +362,6 @@ struct biased_random_walks_functor : public cugraph::c_api::abstract_functor {
 
 struct node2vec_random_walks_functor : public cugraph::c_api::abstract_functor {
   raft::handle_t const& handle_;
-  //  FIXME: rng_state_ should be passed as a parameter
   cugraph::c_api::cugraph_rng_state_t* rng_state_{nullptr};
   cugraph::c_api::cugraph_graph_t* graph_{nullptr};
   cugraph::c_api::cugraph_type_erased_device_array_view_t const* start_vertices_{nullptr};
@@ -375,13 +371,15 @@ struct node2vec_random_walks_functor : public cugraph::c_api::abstract_functor {
   cugraph::c_api::cugraph_random_walk_result_t* result_{nullptr};
 
   node2vec_random_walks_functor(cugraph_resource_handle_t const* handle,
+                                cugraph_rng_state_t* rng_state,
                                 cugraph_graph_t* graph,
                                 cugraph_type_erased_device_array_view_t const* start_vertices,
                                 size_t max_length,
                                 double p,
                                 double q)
     : abstract_functor(),
       handle_(*reinterpret_cast<cugraph::c_api::cugraph_resource_handle_t const*>(handle)->handle_),
+      rng_state_(reinterpret_cast<cugraph::c_api::cugraph_rng_state_t*>(rng_state)),
       graph_(reinterpret_cast<cugraph::c_api::cugraph_graph_t*>(graph)),
       start_vertices_(
         reinterpret_cast<cugraph::c_api::cugraph_type_erased_device_array_view_t const*>(
@@ -403,8 +401,6 @@ struct node2vec_random_walks_functor : public cugraph::c_api::abstract_functor {
     // FIXME: Think about how to handle SG vice MG
     if constexpr (!cugraph::is_candidate<vertex_t, edge_t, weight_t>::value) {
       unsupported();
-    } else if constexpr (multi_gpu) {
-      unsupported();
     } else {
       // random walks expects store_transposed == false
       if constexpr (store_transposed) {
@@ -443,10 +439,6 @@ struct node2vec_random_walks_functor : public cugraph::c_api::abstract_functor {
         graph_view.local_vertex_partition_range_last(),
         false);
 
-      //  FIXME: remove once rng_state passed as parameter
-      rng_state_ = reinterpret_cast<cugraph::c_api::cugraph_rng_state_t*>(
-        new cugraph::c_api::cugraph_rng_state_t{raft::random::RngState{0}});
-
       auto [paths, weights] = cugraph::node2vec_random_walks(
         handle_,
         rng_state_->rng_state_,
@@ -464,8 +456,13 @@ struct node2vec_random_walks_functor : public cugraph::c_api::abstract_functor {
       //
       // Need to unrenumber the vertices in the resulting paths
       //
-      cugraph::unrenumber_local_int_vertices<vertex_t>(
-        handle_, paths.data(), paths.size(), number_map->data(), 0, paths.size(), false);
+      cugraph::unrenumber_int_vertices<vertex_t, multi_gpu>(
+        handle_,
+        paths.data(),
+        paths.size(),
+        number_map->data(),
+        graph_view.vertex_partition_range_lasts(),
+        false);
 
       result_ = new cugraph::c_api::cugraph_random_walk_result_t{
         false,
@@ -546,6 +543,7 @@ void cugraph_random_walk_result_free(cugraph_random_walk_result_t* result)
 
 cugraph_error_code_t cugraph_uniform_random_walks(
   const cugraph_resource_handle_t* handle,
+  cugraph_rng_state_t* rng_state,
   cugraph_graph_t* graph,
   const cugraph_type_erased_device_array_view_t* start_vertices,
   size_t max_length,
@@ -560,13 +558,14 @@ cugraph_error_code_t cugraph_uniform_random_walks(
                "vertex type of graph and start_vertices must match",
                *error);
 
-  uniform_random_walks_functor functor(handle, graph, start_vertices, max_length);
+  uniform_random_walks_functor functor(handle, rng_state, graph, start_vertices, max_length);
 
   return cugraph::c_api::run_algorithm(graph, functor, result, error);
 }
 
 cugraph_error_code_t cugraph_biased_random_walks(
   const cugraph_resource_handle_t* handle,
+  cugraph_rng_state_t* rng_state,
   cugraph_graph_t* graph,
   const cugraph_type_erased_device_array_view_t* start_vertices,
   size_t max_length,
@@ -581,13 +580,14 @@ cugraph_error_code_t cugraph_biased_random_walks(
                "vertex type of graph and start_vertices must match",
                *error);
 
-  biased_random_walks_functor functor(handle, graph, start_vertices, max_length);
+  biased_random_walks_functor functor(handle, rng_state, graph, start_vertices, max_length);
 
   return cugraph::c_api::run_algorithm(graph, functor, result, error);
 }
 
 cugraph_error_code_t cugraph_node2vec_random_walks(
   const cugraph_resource_handle_t* handle,
+  cugraph_rng_state_t* rng_state,
   cugraph_graph_t* graph,
   const cugraph_type_erased_device_array_view_t* start_vertices,
   size_t max_length,
@@ -604,7 +604,7 @@ cugraph_error_code_t cugraph_node2vec_random_walks(
                "vertex type of graph and start_vertices must match",
                *error);
 
-  node2vec_random_walks_functor functor(handle, graph, start_vertices, max_length, p, q);
+  node2vec_random_walks_functor functor(handle, rng_state, graph, start_vertices, max_length, p, q);
 
   return cugraph::c_api::run_algorithm(graph, functor, result, error);
 }