Merge branch 'pyg-dist' of https://github.com/alexbarghi-nv/cugraph i…

…nto pyg-dist
rapidsai · May 29, 2024 · 0cb3d27 · 0cb3d27
2 parents 0551718 + 94cc6d1
commit 0cb3d27
Show file tree

Hide file tree

Showing 3 changed files with 286 additions and 129 deletions.
diff --git a/ci/test_wheel_cugraph-dgl.sh b/ci/test_wheel_cugraph-dgl.sh
@@ -32,8 +32,18 @@ fi
 PYTORCH_URL="https://download.pytorch.org/whl/cu${PYTORCH_CUDA_VER}"
 DGL_URL="https://data.dgl.ai/wheels/cu${PYTORCH_CUDA_VER}/repo.html"
 
+# Starting from 2.2, PyTorch wheels depend on nvidia-nccl-cuxx>=2.19 wheel and
+# dynamically link to NCCL. RAPIDS CUDA 11 CI images have an older NCCL version that
+# might shadow the newer NCCL required by PyTorch during import (when importing
+# `cupy` before `torch`).
+if [[ "${NCCL_VERSION}" < "2.19" ]]; then
+  PYTORCH_VER="2.1.0"
+else
+  PYTORCH_VER="2.3.0"
+fi
+
 rapids-logger "Installing PyTorch and DGL"
-rapids-retry python -m pip install torch --index-url ${PYTORCH_URL}
+rapids-retry python -m pip install "torch==${PYTORCH_VER}" --index-url ${PYTORCH_URL}
 rapids-retry python -m pip install dgl==2.0.0 --find-links ${DGL_URL}
 
 python -m pytest python/cugraph-dgl/tests
diff --git a/docs/cugraph/source/nx_cugraph/nx_cugraph.md b/docs/cugraph/source/nx_cugraph/nx_cugraph.md
@@ -24,142 +24,45 @@ Each chart has three measurements.
 ![Single Source Shortest Path](../images/sssp.png)
 ![Weakly Connected Components](../images/wcc.png)
 
+### Command line example
+Open bc_demo.ipy and paste the code below.
 
-The following algorithms are supported and automatically dispatched to nx-cuGraph for acceleration.
+```
+import pandas as pd
+import networkx as nx
+
+url = "https://data.rapids.ai/cugraph/datasets/cit-Patents.csv"
+df = pd.read_csv(url, sep=" ", names=["src", "dst"], dtype="int32")
+G = nx.from_pandas_edgelist(df, source="src", target="dst")
 
-#### Algorithms
+%time result = nx.betweenness_centrality(G, k=10)
+```
+Run the command:
 ```
-bipartite
- ├─ basic
- │   └─ is_bipartite
- └─ generators
-     └─ complete_bipartite_graph
-centrality
- ├─ betweenness
- │   ├─ betweenness_centrality
- │   └─ edge_betweenness_centrality
- ├─ degree_alg
- │   ├─ degree_centrality
- │   ├─ in_degree_centrality
- │   └─ out_degree_centrality
- ├─ eigenvector
- │   └─ eigenvector_centrality
- └─ katz
-     └─ katz_centrality
-cluster
- ├─ average_clustering
- ├─ clustering
- ├─ transitivity
- └─ triangles
-community
- └─ louvain
-     └─ louvain_communities
-components
- ├─ connected
- │   ├─ connected_components
- │   ├─ is_connected
- │   ├─ node_connected_component
- │   └─ number_connected_components
- └─ weakly_connected
-     ├─ is_weakly_connected
-     ├─ number_weakly_connected_components
-     └─ weakly_connected_components
-core
- ├─ core_number
- └─ k_truss
-dag
- ├─ ancestors
- └─ descendants
-isolate
- ├─ is_isolate
- ├─ isolates
- └─ number_of_isolates
-link_analysis
- ├─ hits_alg
- │   └─ hits
- └─ pagerank_alg
-     └─ pagerank
-operators
- └─ unary
-     ├─ complement
-     └─ reverse
-reciprocity
- ├─ overall_reciprocity
- └─ reciprocity
-shortest_paths
- └─ unweighted
-     ├─ single_source_shortest_path_length
-     └─ single_target_shortest_path_length
-traversal
- └─ breadth_first_search
-     ├─ bfs_edges
-     ├─ bfs_layers
-     ├─ bfs_predecessors
-     ├─ bfs_successors
-     ├─ bfs_tree
-     ├─ descendants_at_distance
-     └─ generic_bfs_edges
-tree
- └─ recognition
-     ├─ is_arborescence
-     ├─ is_branching
-     ├─ is_forest
-     └─ is_tree
+user@machine:/# ipython bc_demo.ipy
 ```
 
-#### Generators
+You will observe a run time of approximately 7 minutes...more or less depending on your cpu.
+
+Run the command again, this time specifiying cugraph as the NetworkX backend of choice.
+```
+user@machine:/# NETWORKX_BACKEND_PRIORITY=cugraph ipython bc_demo.ipy
 ```
-classic
- ├─ barbell_graph
- ├─ circular_ladder_graph
- ├─ complete_graph
- ├─ complete_multipartite_graph
- ├─ cycle_graph
- ├─ empty_graph
- ├─ ladder_graph
- ├─ lollipop_graph
- ├─ null_graph
- ├─ path_graph
- ├─ star_graph
- ├─ tadpole_graph
- ├─ trivial_graph
- ├─ turan_graph
- └─ wheel_graph
-community
- └─ caveman_graph
-small
- ├─ bull_graph
- ├─ chvatal_graph
- ├─ cubical_graph
- ├─ desargues_graph
- ├─ diamond_graph
- ├─ dodecahedral_graph
- ├─ frucht_graph
- ├─ heawood_graph
- ├─ house_graph
- ├─ house_x_graph
- ├─ icosahedral_graph
- ├─ krackhardt_kite_graph
- ├─ moebius_kantor_graph
- ├─ octahedral_graph
- ├─ pappus_graph
- ├─ petersen_graph
- ├─ sedgewick_maze_graph
- ├─ tetrahedral_graph
- ├─ truncated_cube_graph
- ├─ truncated_tetrahedron_graph
- └─ tutte_graph
-social
- ├─ davis_southern_women_graph
- ├─ florentine_families_graph
- ├─ karate_club_graph
- └─ les_miserables_graph
+This run will be much faster, typically around 20 seconds depending on your GPU.
+```
+user@machine:/# NETWORKX_BACKEND_PRIORITY=cugraph ipython bc_demo.ipy
+```
+There is also an option to add caching. This will dramatically help performance when running multiple algorithms on the same graph.
+```
+NETWORKX_BACKEND_PRIORITY=cugraph CACHE_CONVERTED_GRAPH=True ipython bc_demo.ipy
 ```
 
-#### Other
+When running Python interactively, cugraph backend can be specified as an argument in the algorithm call.
 
+For example:
 ```
-convert_matrix
- ├─ from_pandas_edgelist
- └─ from_scipy_sparse_array
+nx.betweenness_centrality(cit_patents_graph, k=k, backend="cugraph")
 ```
+
+
+The latest list of algorithms that can be dispatched to nx-cuGraph for acceleration is found [here](https://github.com/rapidsai/cugraph/blob/main/python/nx-cugraph/README.md#algorithms).