Merge remote-tracking branch 'upstream/branch-24.04' into tests/pytes…

…t/warning_failure
rapidsai · Mar 13, 2024 · f971bcb · f971bcb
2 parents a353e81 + 3086f83
commit f971bcb
Show file tree

Hide file tree

Showing 61 changed files with 3,113 additions and 634 deletions.
diff --git a/benchmarks/nx-cugraph/pytest-based/bench_algos.py b/benchmarks/nx-cugraph/pytest-based/bench_algos.py
@@ -242,6 +242,28 @@ def get_highest_degree_node(graph_obj):
     return max(degrees, key=lambda t: t[1])[0]
 
 
+def build_personalization_dict(pagerank_dict):
+    """
+    Returns a dictionary that can be used as the personalization value for a
+    call to nx.pagerank(). The pagerank_dict passed in is used as the initial
+    source of values for each node, and this function simply treats the list of
+    dict values as two halves (halves A and B) and swaps them so (most if not
+    all) nodes/keys are assigned a different value from the dictionary.
+    """
+    num_half = len(pagerank_dict) // 2
+    A_half_items = list(pagerank_dict.items())[:num_half]
+    B_half_items = list(pagerank_dict.items())[num_half:]
+
+    # Support an odd number of items by initializing with B_half_items, which
+    # will always be one bigger if the number of items is odd. This will leave
+    # the one remainder (in the case of an odd number) unchanged.
+    pers_dict = dict(B_half_items)
+    pers_dict.update({A_half_items[i][0]: B_half_items[i][1] for i in range(num_half)})
+    pers_dict.update({B_half_items[i][0]: A_half_items[i][1] for i in range(num_half)})
+
+    return pers_dict
+
+
 ################################################################################
 # Benchmarks
 def bench_from_networkx(benchmark, graph_obj):
@@ -431,6 +453,26 @@ def bench_pagerank(benchmark, graph_obj, backend_wrapper):
     assert type(result) is dict
 
 
+def bench_pagerank_personalized(benchmark, graph_obj, backend_wrapper):
+    G = get_graph_obj_for_benchmark(graph_obj, backend_wrapper)
+
+    # FIXME: This will run for every combination of inputs, even if the
+    # graph/dataset does not change. Ideally this is run once per
+    # graph/dataset.
+    pagerank_dict = nx.pagerank(G)
+    personalization_dict = build_personalization_dict(pagerank_dict)
+
+    result = benchmark.pedantic(
+        target=backend_wrapper(nx.pagerank),
+        args=(G,),
+        kwargs={"personalization": personalization_dict},
+        rounds=rounds,
+        iterations=iterations,
+        warmup_rounds=warmup_rounds,
+    )
+    assert type(result) is dict
+
+
 def bench_single_source_shortest_path_length(benchmark, graph_obj, backend_wrapper):
     G = get_graph_obj_for_benchmark(graph_obj, backend_wrapper)
     node = get_highest_degree_node(graph_obj)
@@ -804,3 +846,73 @@ def bench_weakly_connected_components(benchmark, graph_obj, backend_wrapper):
         warmup_rounds=warmup_rounds,
     )
     assert type(result) is list
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_complete_bipartite_graph(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_connected_components(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_is_connected(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_node_connected_component(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_number_connected_components(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_is_isolate(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_isolates(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_number_of_isolates(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_complement(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_reverse(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_is_arborescence(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_is_branching(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_is_forest(benchmark, graph_obj, backend_wrapper):
+    pass
+
+
+@pytest.mark.skip(reason="benchmark not implemented")
+def bench_is_tree(benchmark, graph_obj, backend_wrapper):
+    pass
diff --git a/build.sh b/build.sh
@@ -18,7 +18,7 @@ ARGS=$*
 # script, and that this script resides in the repo dir!
 REPODIR=$(cd $(dirname $0); pwd)
 
-RAPIDS_VERSION=24.04
+RAPIDS_VERSION="$(sed -E -e 's/^([0-9]{2})\.([0-9]{2})\.([0-9]{2}).*$/\1.\2/' VERSION)"
 
 # Valid args to this script (all possible targets and options) - only one per line
 VALIDARGS="

diff --git a/ci/build_docs.sh b/ci/build_docs.sh
@@ -39,7 +39,9 @@ rapids-mamba-retry install \
 rapids-logger "Install cugraph-dgl"
 rapids-mamba-retry install "${PYTHON_CHANNEL}/linux-64/cugraph-dgl-*.tar.bz2"
 
-export RAPIDS_VERSION_NUMBER="24.04"
+export RAPIDS_VERSION="$(rapids-version)"
+export RAPIDS_VERSION_MAJOR_MINOR="$(rapids-version-major-minor)"
+export RAPIDS_VERSION_NUMBER="$RAPIDS_VERSION_MAJOR_MINOR"
 export RAPIDS_DOCS_DIR="$(mktemp -d)"
 
 for PROJECT in libcugraphops libwholegraph; do

diff --git a/ci/build_python.sh b/ci/build_python.sh
@@ -23,10 +23,11 @@ echo "${version}" > VERSION
 rapids-logger "Begin py build"
 
 package_dir="python"
-for package_name in pylibcugraph cugraph nx-cugraph cugraph-pyg cugraph-dgl; do
+for package_name in pylibcugraph cugraph cugraph-pyg cugraph-dgl; do
   underscore_package_name=$(echo "${package_name}" | tr "-" "_")
   sed -i "/^__git_commit__/ s/= .*/= \"${git_commit}\"/g" "${package_dir}/${package_name}/${underscore_package_name}/_version.py"
 done
+sed -i "/^__git_commit__/ s/= .*/= \"${git_commit}\"/g" "${package_dir}/nx-cugraph/_nx_cugraph/_version.py"
 
 # TODO: Remove `--no-test` flags once importing on a CPU
 # node works correctly

diff --git a/ci/build_wheel.sh b/ci/build_wheel.sh
@@ -20,8 +20,12 @@ RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
 PACKAGE_CUDA_SUFFIX="-${RAPIDS_PY_CUDA_SUFFIX}"
 
 # Patch project metadata files to include the CUDA version suffix and version override.
+version_package_name="$underscore_package_name"
+if [[ "${version_package_name}" = "nx_cugraph" ]]; then
+    version_package_name="_nx_cugraph"
+fi
 pyproject_file="${package_dir}/pyproject.toml"
-version_file="${package_dir}/${underscore_package_name}/_version.py"
+version_file="${package_dir}/${version_package_name}/_version.py"
 
 sed -i "s/name = \"${package_name}\"/name = \"${package_name}${PACKAGE_CUDA_SUFFIX}\"/g" ${pyproject_file}
 echo "${version}" > VERSION

diff --git a/ci/release/update-version.sh b/ci/release/update-version.sh
@@ -39,33 +39,11 @@ function sed_runner() {
     sed -i.bak ''"$1"'' $2 && rm -f ${2}.bak
 }
 
-# rapids-cmake version
-sed_runner 's/'"branch-.*\/RAPIDS.cmake"'/'"branch-${NEXT_SHORT_TAG}\/RAPIDS.cmake"'/g' fetch_rapids.cmake
-
-# CMakeLists update
-sed_runner 's/'"CUGRAPH VERSION .* LANGUAGES C CXX CUDA)"'/'"CUGRAPH VERSION ${NEXT_FULL_TAG} LANGUAGES C CXX CUDA)"'/g' cpp/CMakeLists.txt
-sed_runner 's|'"branch-.*/RAPIDS.cmake"'|'"branch-${NEXT_SHORT_TAG}/RAPIDS.cmake"'|g' cpp/CMakeLists.txt
-sed_runner 's/'"CUGRAPH_ETL VERSION .* LANGUAGES C CXX CUDA)"'/'"CUGRAPH_ETL VERSION ${NEXT_FULL_TAG} LANGUAGES C CXX CUDA)"'/g' cpp/libcugraph_etl/CMakeLists.txt
-sed_runner 's|'"branch-.*/RAPIDS.cmake"'|'"branch-${NEXT_SHORT_TAG}/RAPIDS.cmake"'|g' cpp/libcugraph_etl/CMakeLists.txt
-sed_runner "s/set(pylibcugraph_version .*)/set(pylibcugraph_version ${NEXT_FULL_TAG})/g" python/pylibcugraph/CMakeLists.txt
-sed_runner "s/set(cugraph_version .*)/set(cugraph_version ${NEXT_FULL_TAG})/g" python/cugraph/CMakeLists.txt
-
-# RTD update
-sed_runner 's/version = .*/version = '"'${NEXT_SHORT_TAG}'"'/g' docs/cugraph/source/conf.py
-sed_runner 's/release = .*/release = '"'${NEXT_FULL_TAG}'"'/g' docs/cugraph/source/conf.py
-
-
-# build.sh script
-sed_runner 's/RAPIDS_VERSION=.*/RAPIDS_VERSION='${NEXT_SHORT_TAG}'/g' build.sh
-
 # Centralized version file update
 # NOTE: Any script that runs in CI will need to use gha-tool `rapids-generate-version`
 # and echo it to `VERSION` file to get an alpha spec of the current version
 echo "${NEXT_FULL_TAG}" > VERSION
 
-# Wheel testing script
-sed_runner "s/branch-.*/branch-${NEXT_SHORT_TAG}/g" ci/test_wheel_cugraph.sh
-
 # Need to distutils-normalize the original version
 NEXT_SHORT_TAG_PEP440=$(python -c "from setuptools.extern import packaging; print(packaging.version.Version('${NEXT_SHORT_TAG}'))")
 
@@ -105,19 +83,11 @@ for DEP in "${DEPENDENCIES[@]}"; do
   done
 done
 
-# Doxyfile update
-sed_runner "s|PROJECT_NUMBER[[:space:]]*=.*|PROJECT_NUMBER=${NEXT_SHORT_TAG}|" cpp/doxygen/Doxyfile
-
 # ucx-py version
 sed_runner "/^ucx_py_version:$/ {n;s/.*/  - \"${NEXT_UCX_PY_VERSION}.*\"/}" conda/recipes/cugraph/conda_build_config.yaml
 sed_runner "/^ucx_py_version:$/ {n;s/.*/  - \"${NEXT_UCX_PY_VERSION}.*\"/}" conda/recipes/cugraph-service/conda_build_config.yaml
 sed_runner "/^ucx_py_version:$/ {n;s/.*/  - \"${NEXT_UCX_PY_VERSION}.*\"/}" conda/recipes/pylibcugraph/conda_build_config.yaml
 
-# nx-cugraph NetworkX entry-point meta-data
-sed_runner "s@branch-[0-9][0-9].[0-9][0-9]@branch-${NEXT_SHORT_TAG}@g" python/nx-cugraph/_nx_cugraph/__init__.py
-# FIXME: can this use the standard VERSION file and update mechanism?
-sed_runner "s/__version__ = .*/__version__ = \"${NEXT_FULL_TAG}\"/g" python/nx-cugraph/_nx_cugraph/__init__.py
-
 # CI files
 for FILE in .github/workflows/*.yaml; do
   sed_runner "/shared-workflows/ s/@.*/@branch-${NEXT_SHORT_TAG}/g" "${FILE}"
@@ -126,7 +96,6 @@ for FILE in .github/workflows/*.yaml; do
   # Wheel builds install dask-cuda from source, update its branch
   sed_runner "s/dask-cuda.git@branch-[0-9][0-9].[0-9][0-9]/dask-cuda.git@branch-${NEXT_SHORT_TAG}/g" "${FILE}"
 done
-sed_runner "s/RAPIDS_VERSION_NUMBER=\".*/RAPIDS_VERSION_NUMBER=\"${NEXT_SHORT_TAG}\"/g" ci/build_docs.sh
 
 sed_runner "s/branch-.*/branch-${NEXT_SHORT_TAG}/g" python/nx-cugraph/README.md
 

diff --git a/conda/environments/all_cuda-118_arch-x86_64.yaml b/conda/environments/all_cuda-118_arch-x86_64.yaml
@@ -42,7 +42,7 @@ dependencies:
 - ninja
 - notebook>=0.5.0
 - numba>=0.57
-- numpy>=1.23
+- numpy>=1.23,<2.0a0
 - numpydoc
 - nvcc_linux-64=11.8
 - openmpi

diff --git a/conda/environments/all_cuda-122_arch-x86_64.yaml b/conda/environments/all_cuda-122_arch-x86_64.yaml
@@ -48,7 +48,7 @@ dependencies:
 - ninja
 - notebook>=0.5.0
 - numba>=0.57
-- numpy>=1.23
+- numpy>=1.23,<2.0a0
 - numpydoc
 - openmpi
 - packaging>=21

diff --git a/conda/recipes/cugraph-dgl/meta.yaml b/conda/recipes/cugraph-dgl/meta.yaml
@@ -25,7 +25,7 @@ requirements:
     - cugraph ={{ version }}
     - dgl >=1.1.0.cu*
     - numba >=0.57
-    - numpy >=1.23
+    - numpy >=1.23,<2.0a0
     - pylibcugraphops ={{ minor_version }}
     - python
     - pytorch

diff --git a/conda/recipes/cugraph-pyg/meta.yaml b/conda/recipes/cugraph-pyg/meta.yaml
@@ -28,7 +28,7 @@ requirements:
   run:
     - rapids-dask-dependency ={{ minor_version }}
     - numba >=0.57
-    - numpy >=1.23
+    - numpy >=1.23,<2.0a0
     - python
     - pytorch >=2.0
     - cupy >=12.0.0

diff --git a/conda/recipes/cugraph-service/meta.yaml b/conda/recipes/cugraph-service/meta.yaml
@@ -60,7 +60,7 @@ outputs:
         - dask-cuda ={{ minor_version }}
         - dask-cudf ={{ minor_version }}
         - numba >=0.57
-        - numpy >=1.23
+        - numpy >=1.23,<2.0a0
         - python
         - rapids-dask-dependency ={{ minor_version }}
         - thriftpy2 >=0.4.15

diff --git a/cpp/CMakeLists.txt b/cpp/CMakeLists.txt
@@ -16,7 +16,7 @@
 
 cmake_minimum_required(VERSION 3.26.4 FATAL_ERROR)
 
-include(../fetch_rapids.cmake)
+include(../rapids_config.cmake)
 include(rapids-cmake)
 include(rapids-cpm)
 include(rapids-cuda)
@@ -25,7 +25,7 @@ include(rapids-find)
 
 rapids_cuda_init_architectures(CUGRAPH)
 
-project(CUGRAPH VERSION 24.04.00 LANGUAGES C CXX CUDA)
+project(CUGRAPH VERSION "${RAPIDS_VERSION}" LANGUAGES C CXX CUDA)
 
 if(CMAKE_CUDA_COMPILER_ID STREQUAL "NVIDIA" AND
    CMAKE_CUDA_COMPILER_VERSION VERSION_LESS 11.0)
@@ -590,7 +590,7 @@ find_package(Doxygen 1.8.11)
 if(Doxygen_FOUND)
     add_custom_command(OUTPUT CUGRAPH_DOXYGEN
                        WORKING_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}/doxygen
-                       COMMAND doxygen Doxyfile
+                       COMMAND ${CMAKE_COMMAND} -E env "RAPIDS_VERSION_MAJOR_MINOR=${RAPIDS_VERSION_MAJOR_MINOR}" doxygen Doxyfile
                        VERBATIM)
 
     add_custom_target(docs_cugraph DEPENDS CUGRAPH_DOXYGEN)

diff --git a/cpp/doxygen/Doxyfile b/cpp/doxygen/Doxyfile
@@ -48,7 +48,7 @@ PROJECT_NAME           = libcugraph
 # could be handy for archiving the generated documentation or if some version
 # control system is used.
 
-PROJECT_NUMBER=24.04
+PROJECT_NUMBER=$(RAPIDS_VERSION_MAJOR_MINOR)
 
 # Using the PROJECT_BRIEF tag one can provide an optional one line description
 # for a project that appears at the top of each page and should give viewer a

diff --git a/cpp/include/cugraph/graph_view.hpp b/cpp/include/cugraph/graph_view.hpp
@@ -613,7 +613,7 @@ class graph_view_t<vertex_t, edge_t, store_transposed, multi_gpu, std::enable_if
       major_value_range_start_offset);
   }
 
-  // FIXME: deprecated, replaced with copmute_number_of_edges (which works with or without edge
+  // FIXME: deprecated, replaced with compute_number_of_edges (which works with or without edge
   // masking)
   edge_t number_of_edges() const
   {
@@ -923,7 +923,7 @@ class graph_view_t<vertex_t, edge_t, store_transposed, multi_gpu, std::enable_if
       offsets_, indices_, this->number_of_vertices());
   }
 
-  // FIXME: deprecated, replaced with copmute_number_of_edges (which works with or without edge
+  // FIXME: deprecated, replaced with compute_number_of_edges (which works with or without edge
   // masking)
   edge_t number_of_edges() const
   {

diff --git a/cpp/include/cugraph/utilities/misc_utils.cuh b/cpp/include/cugraph/utilities/misc_utils.cuh
@@ -94,6 +94,14 @@ thrust::optional<T> to_thrust_optional(std::optional<T> val)
   return ret;
 }
 
+template <typename T>
+std::optional<T> to_std_optional(thrust::optional<T> val)
+{
+  std::optional<T> ret{std::nullopt};
+  if (val) { ret = *val; }
+  return ret;
+}
+
 template <typename idx_t, typename offset_t>
 rmm::device_uvector<idx_t> expand_sparse_offsets(raft::device_span<offset_t const> offsets,
                                                  idx_t base_idx,

diff --git a/cpp/libcugraph_etl/CMakeLists.txt b/cpp/libcugraph_etl/CMakeLists.txt
@@ -15,7 +15,7 @@
 #=============================================================================
 
 cmake_minimum_required(VERSION 3.26.4 FATAL_ERROR)
-include(../../fetch_rapids.cmake)
+include(../../rapids_config.cmake)
 
 include(rapids-cmake)
 include(rapids-cpm)
@@ -25,7 +25,7 @@ include(rapids-find)
 
 rapids_cuda_init_architectures(CUGRAPH_ETL)
 
-project(CUGRAPH_ETL VERSION 24.04.00 LANGUAGES C CXX CUDA)
+project(CUGRAPH_ETL VERSION "${RAPIDS_VERSION}" LANGUAGES C CXX CUDA)
 
 if(CMAKE_CUDA_COMPILER_ID STREQUAL "NVIDIA" AND
    CMAKE_CUDA_COMPILER_VERSION VERSION_LESS 11.0)