Releases: rapidsai/cugraph
Releases · rapidsai/cugraph
v24.08.00
🚨 Breaking Changes
🐛 Bug Fixes
- Fix ucx-py update-version script (#4603) @raydouglass
- add setuptools to host requirements for conda packages that need it (#4582) @jameslamb
- Add pylibcugraph dependency on pylibraft. (#4570) @bdice
- Fix build error with NO_CUGRAPH_OPS (#4563) @seunghwak
- [BUG] Fix Failing WholeGraph Tests (#4560) @alexbarghi-nv
- Temporarily Disable Feature Store Tests with WholeGraph (#4559) @alexbarghi-nv
- Fix MG
katz_centrality
: Check if DataFrame Arg is Not None (#4555) @nv-rliu - nx-cugraph: fix
from_pandas_edgekey
given edgekey but not edgeattr (#4550) @eriknw - Fix triangle count test bug (#4549) @jnke2016
- [BUG] Use the Correct WG Communicator (#4548) @alexbarghi-nv
- Add Additional Check For SSSP Source Vertex & Fix SSSP Benchmark (#4541) @nv-rliu
- Fix OOM Bug for Jaccard, Sorensen, and Overlap benchmarks (#4524) @nv-rliu
- Distribute start_list across ranks (#4519) @jnke2016
- [FIX] Skip Distributed Sampler Tests if PyTorch with CUDA is not Available (#4518) @alexbarghi-nv
- [BUG] Fix a hang issue in MG triangle counts (when invoked with a small number of vertices to update triangle counts) (#4517) @seunghwak
- Update MG Benchmark List (#4516) @nv-rliu
- Fix TensorProductConv test and improve docs (#4480) @tingyu66
- Test nx-cugraph package instead of editable install (#4442) @KyleFromNVIDIA
📖 Documentation
- DOC: typo in nx_transition.rst (#4491) @raybellwaves
- Doc cleanup for nx-cugraph: fixed typos, cleaned up various descriptions, renamed notebook to match naming convetion. (#4478) @rlratzel
- [DOC] Minor Improvements to cuGraph-PyG Documentation (#4460) @alexbarghi-nv
🚀 New Features
- Use MNMG version of ECG in python layer instead, and remove legacy ECG and Louvain (#4514) @naimnv
- c_api and plc binding for lookup src dst using edge ids and type(s) (#4494) @naimnv
- Forward merge branch-24.06 into branch-24.08 (#4489) @nv-rliu
- [FEA] New Graph Interface and Loaders for Distributed Sampling in DGL (#4486) @alexbarghi-nv
- compute cosine similarity for vertex pairs (#4482) @naimnv
- Define heterogeneous renumbering API (#4463) @seunghwak
- Lookup edge src dst using edge id and type (#4449) @naimnv
- Biased sampling (#4443) @seunghwak
🛠️ Improvements
- nx-cugraph: check networkx version (#4571) @eriknw
- nx-cugraph: add
G.__networkx_cache__
to enable graph conversion caching (#4567) @eriknw - split up CUDA-suffixed dependencies in dependencies.yaml (#4552) @jameslamb
- Use workflow branch 24.08 again (#4544) @KyleFromNVIDIA
- Support non p2p configuration when initializing the comms (#4543) @jnke2016
- Fix Warning from
simpleDistributedGraph.py
(#4540) @nv-rliu - Create a graph from the edge list in multiple chunks (#4539) @seunghwak
- nx-cugraph: add dijkstra sssp functions (#4538) @eriknw
- nx-cugraph: add
from_dict_of_lists
andto_dict_of_lists
(#4537) @eriknw - Ensure
get_test_data.sh
doesn't re-download datasets (#4536) @trxcllnt - Define and Implement C API for biased sampling (#4535) @ChuckHastings
- Build and test with CUDA 12.5.1 (#4534) @KyleFromNVIDIA
- Refactor C++ unit tests to allow finer grained filtering (#4533) @ChuckHastings
- [IMP] Set the Default WG Memory Type to 'distributed' for the MNMG PyG Example (#4532) @alexbarghi-nv
- nx-cugraph: add
relabel_nodes
andconvert_node_labels_to_integers
(#4531) @eriknw - Add
-cuXX
suffixed versions of cugraph-service-client dependency to pyproject.toml's project.dependencies list (#4530) @trxcllnt - Further optimize
from_pandas_edgelist
with cudf (#4528) @eriknw - Performance optimize BFS (including direction optimizing BFS implementation, mainly for single-GPU) (#4527) @seunghwak
- Add CUDA_STATIC_MATH_LIBRARIES (#4526) @KyleFromNVIDIA
- Better handle cudf.pandas in
from_pandas_edgelist
(#4525) @eriknw - Skip the benchmark ctests within CI (#4522) @ChuckHastings
- remove thriftpy2 ceiling (#4521) @jameslamb
- Avoid --find-links in wheel jobs (#4509) @jameslamb
- Refactor code base to reduce memory requirement for building libcugraph (#4506) @naimnv
- Tweak rmm configuration for C++ unit tests (#4503) @ChuckHastings
- Expose new all-pairs Similarity algorithms (#4502) @jnke2016
- remove openmpi ceiling (#4496) @jameslamb
- Cut peak memory footprint in per_v_transform_reduce_dst_key_aggregated_outgoing_e (#4484) @seunghwak
- Skip MG
dgl_uniform_sampler
test in nightlies (#4479) @nv-rliu - Remove text builds of documentation (#4468) @vyasr
- [IMP] Limit the Test Data Size when Running CI in
gcn_dist_sg.py
(#4461) @alexbarghi-nv - Forward Merge branch-24.06 into branch-24.08 (#4454) @nv-rliu
- Properly clean up python directories (#4453) @ChuckHastings
- Fixes for On-Going MG Test Failures (#4450) @nv-rliu
- remove unnecessary 'setuptools' and 'wheel' dependencies (#4448) @jameslamb
- MG Implementation K-Truss (#4438) @jnke2016
- Overhaul ops-codeowners (#4409) @raydouglass
- Use rapids-build-backend (#4393) @KyleFromNVIDIA
- Optimize K-Truss (#4375) @jnke2016
[NIGHTLY] v24.10.00
🔗 Links
🚨 Breaking Changes
- Add Additional Check for
NetworkX
Release Candidate Versions (#4613) @nv-rliu - Updates to
cugraph.hypergraph
(Duplicate Col Labels Bug) (#4610) @nv-rliu - Heterogeneous renumbering implementation (#4602) @seunghwak
- [FEA] DGL Examples (#4583) @alexbarghi-nv
🐛 Bug Fixes
- Fix build strings in nx-cugraph (#4639) @bdice
- Set CUDA_STATIC_MATH_LIBRARIES in Python builds (#4612) @KyleFromNVIDIA
- Updates to
cugraph.hypergraph
(Duplicate Col Labels Bug) (#4610) @nv-rliu - Biased sampling primitive bug fix (#4607) @seunghwak
- Fix
test_property_graph_mg
Usage of Util Function (#4600) @nv-rliu - Re-configure benchmarking devices & add markers to
bench_cugraph_uniform_neighbor_sample
(#4561) @nv-rliu
🚀 New Features
- Heterogeneous renumbering implementation (#4602) @seunghwak
- [FEA] Biased Sampling in cuGraph-PyG (#4586) @alexbarghi-nv
- [FEA] DGL Examples (#4583) @alexbarghi-nv
🛠️ Improvements
- update a few more Python references to Python 3.10 (#4637) @jameslamb
- Negative Sampling test needs whole GPU (#4636) @ChuckHastings
- Update rapidsai/pre-commit-hooks (#4633) @KyleFromNVIDIA
- Update examples to build with latest changes to cugraph (#4632) @ChuckHastings
- Remove Warnings and Timeout from
bench_cugraph_uniform_neighbor_sample.py
(#4631) @nv-rliu - Drop Python 3.9 support (#4625) @jameslamb
- Download fewer datasets for C/C++ unit tests (#4624) @ChuckHastings
- Use CUDA math wheels (#4621) @KyleFromNVIDIA
- Fix ListColumn constructor argument (#4620) @mroeschke
- Use CategoricalColumn instead of build_categorical_column (#4618) @mroeschke
- Add
nx-cugraph
Benchmarking Scripts (#4616) @nv-rliu - Add Additional Check for
NetworkX
Release Candidate Versions (#4613) @nv-rliu - Remove a bunch of legacy code that's no longer used (#4609) @ChuckHastings
- Update pre-commit hooks (#4605) @KyleFromNVIDIA
- Improve update-version.sh (#4599) @bdice
- Use tool.scikit-build.cmake.version, set scikit-build-core minimum-version (#4597) @jameslamb
- Migrate get_sampling_index function from cugraph-ops to cugraph (#4594) @ChuckHastings
- Merge branch-24.08 into branch-24.10 (#4565) @jameslamb
- Fix ucx-py version, use UCX 1.17.0 in pip devcontainers (#4562) @bdice
- Use stream_allocator_adaptor constructor instead of factory. (#4557) @bdice
- Add an Explanatory Error Message for uint Types (#4556) @alexbarghi-nv
- Define and Implement C++ API for negative sampling (#4523) @ChuckHastings
v24.06.01
🚨 Breaking Changes
- Call New
replicate_edgelist
Function (#4441) @nv-rliu - Enable FutureWarnings and DeprecationWarnings as errors in cugraph (#4415) @mroeschke
- Distributed Sampling in cuGraph-PyG (#4384) @alexbarghi-nv
- Move edge triangle count to the stable API (#4382) @jnke2016
- Address dask_cudf.read_csv chunksize deprecation (#4379) @mroeschke
- Convert device_memory_resource* to device_async_resource_ref (#4365) @harrism
- Update cugraph-pyg models for PyG 2.5 (#4335) @tingyu66
🐛 Bug Fixes
- Fix Broken
cpp_build
by Pinning Openmpi Build Ver (24.06) (#4475) @nv-rliu - [BUG] Pin Test Version of PyTorch to 2.1 to Resolve NCCL Error (#4464) @alexbarghi-nv
- Fix bug in kv_store_t's insertion methods (#4444) @naimnv
- Fix a bug in kv_store_t implementation (#4434) @naimnv
- [FIX] Remove Unintentional Expensive Check from SG Uniform Neighbor Sample (#4421) @alexbarghi-nv
- BFS fix for exec_policy (#4417) @ChuckHastings
- Skip/adjust doctests for deprecations (#4380) @mroeschke
- Address dask_cudf.read_csv chunksize deprecation (#4379) @mroeschke
- Fix bad Leiden result (#4376) @ChuckHastings
- Erdos-Renyi generator had bad logic in thrust calls (#4362) @ChuckHastings
- Change mamba option --force to --yes (#4357) @naimnv
- Fix false C++ HITS test failure on H100 (#4352) @seunghwak
- Bug fix in cugraph::detail::extract_transform_v_froniter_e (#4320) @seunghwak
- Refactor CMakeLists.txt to better express usage requirements (#4309) @robertmaynard
- [BUG] Always Persist Dask DataFrames in cuGraph-DGL Graph Storage (#4296) @alexbarghi-nv
- Fix Bug in MG
ego_graph
(#4262) @nv-rliu - [FIX] Remove Unnecessary CUDA Check Causing Deprecation Warning (#4239) @alexbarghi-nv
📖 Documentation
- Fixed links and added c++ docs per issue 4431 (#4435) @acostadon
- first copy of general cugraph tutorial. (#4396) @acostadon
- adding notebook to demo nx_cugraph (#4366) @acostadon
- Update operators.rst (#4339) @raybellwaves
- Removed obsolete methods from rst file (#4331) @acostadon
- Update DGL_support.md (#4327) @raybellwaves
- test sphinx mapping to networkx (#4323) @raybellwaves
- Updates the nx-cugraph README.md with the latest upstream URLs (#4321) @rlratzel
- Fix Table for Synthetic Data notebook (#4311) @nv-rliu
- Fix Docstring for
from_dask_cudf_edgelist
(#4310) @nv-rliu - fix PyPI URL in docs (#4282) @jameslamb
- DOC: doc-update-link-for-cugraphops (#4279) @raybellwaves
- Notebook fixes to formulas and outputs (#4272) @acostadon
🚀 New Features
- [FEA] New WholeGraph Feature Store for PyG (#4432) @alexbarghi-nv
- Biased sampling primitive (#4430) @seunghwak
- Distributed Sampling in cuGraph-PyG (#4384) @alexbarghi-nv
- [FEA] Support Seed Retention for Sampling with Renumbering (#4355) @alexbarghi-nv
- Always use a static gtest (#4344) @robertmaynard
- Update GNN sampling post processing functions to optionally consider seed vertices in renumbering (#4329) @seunghwak
- MNMG Approximation Algorithm for the Weighted Matching Problem (#4315) @naimnv
- [FEA] cuGraph GNN NCCL-only Setup and Distributed Sampling (#4278) @alexbarghi-nv
🛠️ Improvements
- Adds benchmark for
nx_cugraph.ego_graph
(#4451) @rlratzel - Pin torch version in
cugraph-dgl
wheel test (#4447) @tingyu66 - Call New
replicate_edgelist
Function (#4441) @nv-rliu - Fix MG Betweenness Centrality Test Bug (#4429) @nv-rliu
- Expose graph_view method to count multi edges through C API and PLC (#4426) @ChuckHastings
- Update doxygen comments to specify in,out instead of in/out (#4425) @ChuckHastings
- Enable FutureWarnings and DeprecationWarnings as errors in cugraph (#4415) @mroeschke
- Add MTMG support for Similarity and test for all-pairs Jaccard (#4411) @ChuckHastings
- Update collect_comm to handle value of tuple type (#4410) @naimnv
- Updates SG
PropertyGraph
andcugraph-service
to applyDataFrame.fillna()
based on latest cuDF changes (#4408) @rlratzel - Fix building cugraph with CCCL main (#4404) @trxcllnt
- Update decompress_to_edgelist to handle edge types (#4397) @naimnv
- nx-cugraph: add
ego_graph
(#4395) @eriknw - Remove unused benchmark fixture in test_dask_mg_random_walks (#4391) @mroeschke
- Address do_expensive_check FutureWarnings in Python tests (#4390) @mroeschke
- Address k_truss_subgraph(use_weights=) deprecation (#4389) @mroeschke
- Address PendingDeprecationWarning from _from_edgelist (#4388) @mroeschke
- Adjust deprecated cugraph.subgraph usage in Python tests (#4386) @mroeschke
- Update
k_truss
to work with many connected components (#4383) @eriknw - Move edge triangle count to the stable API (#4382) @jnke2016
- Replace deprecated cupy.sparse with cupyx.scipy.sparse (#4381) @mroeschke
- Migrate to
{{ stdlib("c") }}
(#4374) @hcho3 - Convert device_memory_resource* to device_async_resource_ref (#4365) @harrism
- add --rm and --name to devcontainer run args (#4361) @trxcllnt
- Update pip devcontainers to UCX v1.15.0 (#4360) @trxcllnt
- Add missing
<optional>
include (#4353) @trxcllnt nx-cugraph
: supportshould_run
that was added in NetworkX 3.3 (#4348) @eriknw- Update generation and flattening of dendrogram in Leiden (#4347) @ChuckHastings
- Update cugraph-pyg models for PyG 2.5 (#4335) @tingyu66
- Update pre-commit hooks to include all C and C++ files (#4332) @ChuckHastings
- Enable expression-based Dask Dataframe support (#4325) @rjzamora
- Use appropriate edge src and dst properties based on is_multi_gpu flag (#4319) @naimnv
- Test GNN Examples and Add CUDA 12 Testing (#4317) @alexbarghi-nv
- Adds missing files to
update-version.sh
(#4314) @AyodeAwe - Fix stream synchronization in MTMG graph construction (#4275) @ChuckHastings
- Get rid of sentinel namespace (#4274) @PointKernel
- Update more C++ algorithm tests to include edge masked input graphs. (#4267) @seunghwak
- Update triangle count to internally use edge masking and update Triangle counting & K-hop neighbor tests to include edge masked graph view inputs (#4261) @seunghwak
- Add Additional Checks to
get_edgelist
andget_dask_edgelist
(#4256) @nv-rliu - Refactor Several MG Tests (#4244) @nv-rliu
v24.06.00
🚨 Breaking Changes
- Call New
replicate_edgelist
Function (#4441) @nv-rliu - Enable FutureWarnings and DeprecationWarnings as errors in cugraph (#4415) @mroeschke
- Distributed Sampling in cuGraph-PyG (#4384) @alexbarghi-nv
- Move edge triangle count to the stable API (#4382) @jnke2016
- Address dask_cudf.read_csv chunksize deprecation (#4379) @mroeschke
- Convert device_memory_resource* to device_async_resource_ref (#4365) @harrism
- Update cugraph-pyg models for PyG 2.5 (#4335) @tingyu66
🐛 Bug Fixes
- Fix bug in kv_store_t's insertion methods (#4444) @naimnv
- Fix a bug in kv_store_t implementation (#4434) @naimnv
- [FIX] Remove Unintentional Expensive Check from SG Uniform Neighbor Sample (#4421) @alexbarghi-nv
- BFS fix for exec_policy (#4417) @ChuckHastings
- Skip/adjust doctests for deprecations (#4380) @mroeschke
- Address dask_cudf.read_csv chunksize deprecation (#4379) @mroeschke
- Fix bad Leiden result (#4376) @ChuckHastings
- Erdos-Renyi generator had bad logic in thrust calls (#4362) @ChuckHastings
- Change mamba option --force to --yes (#4357) @naimnv
- Fix false C++ HITS test failure on H100 (#4352) @seunghwak
- Bug fix in cugraph::detail::extract_transform_v_froniter_e (#4320) @seunghwak
- Refactor CMakeLists.txt to better express usage requirements (#4309) @robertmaynard
- [BUG] Always Persist Dask DataFrames in cuGraph-DGL Graph Storage (#4296) @alexbarghi-nv
- Fix Bug in MG
ego_graph
(#4262) @nv-rliu - [FIX] Remove Unnecessary CUDA Check Causing Deprecation Warning (#4239) @alexbarghi-nv
📖 Documentation
- Fixed links and added c++ docs per issue 4431 (#4435) @acostadon
- first copy of general cugraph tutorial. (#4396) @acostadon
- adding notebook to demo nx_cugraph (#4366) @acostadon
- Update operators.rst (#4339) @raybellwaves
- Removed obsolete methods from rst file (#4331) @acostadon
- Update DGL_support.md (#4327) @raybellwaves
- test sphinx mapping to networkx (#4323) @raybellwaves
- Updates the nx-cugraph README.md with the latest upstream URLs (#4321) @rlratzel
- Fix Table for Synthetic Data notebook (#4311) @nv-rliu
- Fix Docstring for
from_dask_cudf_edgelist
(#4310) @nv-rliu - fix PyPI URL in docs (#4282) @jameslamb
- DOC: doc-update-link-for-cugraphops (#4279) @raybellwaves
- Notebook fixes to formulas and outputs (#4272) @acostadon
🚀 New Features
- [FEA] New WholeGraph Feature Store for PyG (#4432) @alexbarghi-nv
- Biased sampling primitive (#4430) @seunghwak
- Distributed Sampling in cuGraph-PyG (#4384) @alexbarghi-nv
- [FEA] Support Seed Retention for Sampling with Renumbering (#4355) @alexbarghi-nv
- Always use a static gtest (#4344) @robertmaynard
- Update GNN sampling post processing functions to optionally consider seed vertices in renumbering (#4329) @seunghwak
- MNMG Approximation Algorithm for the Weighted Matching Problem (#4315) @naimnv
- [FEA] cuGraph GNN NCCL-only Setup and Distributed Sampling (#4278) @alexbarghi-nv
🛠️ Improvements
- Adds benchmark for
nx_cugraph.ego_graph
(#4451) @rlratzel - Pin torch version in
cugraph-dgl
wheel test (#4447) @tingyu66 - Call New
replicate_edgelist
Function (#4441) @nv-rliu - Fix MG Betweenness Centrality Test Bug (#4429) @nv-rliu
- Expose graph_view method to count multi edges through C API and PLC (#4426) @ChuckHastings
- Update doxygen comments to specify in,out instead of in/out (#4425) @ChuckHastings
- Enable FutureWarnings and DeprecationWarnings as errors in cugraph (#4415) @mroeschke
- Add MTMG support for Similarity and test for all-pairs Jaccard (#4411) @ChuckHastings
- Update collect_comm to handle value of tuple type (#4410) @naimnv
- Updates SG
PropertyGraph
andcugraph-service
to applyDataFrame.fillna()
based on latest cuDF changes (#4408) @rlratzel - Fix building cugraph with CCCL main (#4404) @trxcllnt
- Update decompress_to_edgelist to handle edge types (#4397) @naimnv
- nx-cugraph: add
ego_graph
(#4395) @eriknw - Remove unused benchmark fixture in test_dask_mg_random_walks (#4391) @mroeschke
- Address do_expensive_check FutureWarnings in Python tests (#4390) @mroeschke
- Address k_truss_subgraph(use_weights=) deprecation (#4389) @mroeschke
- Address PendingDeprecationWarning from _from_edgelist (#4388) @mroeschke
- Adjust deprecated cugraph.subgraph usage in Python tests (#4386) @mroeschke
- Update
k_truss
to work with many connected components (#4383) @eriknw - Move edge triangle count to the stable API (#4382) @jnke2016
- Replace deprecated cupy.sparse with cupyx.scipy.sparse (#4381) @mroeschke
- Migrate to
{{ stdlib("c") }}
(#4374) @hcho3 - Convert device_memory_resource* to device_async_resource_ref (#4365) @harrism
- add --rm and --name to devcontainer run args (#4361) @trxcllnt
- Update pip devcontainers to UCX v1.15.0 (#4360) @trxcllnt
- Add missing
<optional>
include (#4353) @trxcllnt nx-cugraph
: supportshould_run
that was added in NetworkX 3.3 (#4348) @eriknw- Update generation and flattening of dendrogram in Leiden (#4347) @ChuckHastings
- Update cugraph-pyg models for PyG 2.5 (#4335) @tingyu66
- Update pre-commit hooks to include all C and C++ files (#4332) @ChuckHastings
- Enable expression-based Dask Dataframe support (#4325) @rjzamora
- Use appropriate edge src and dst properties based on is_multi_gpu flag (#4319) @naimnv
- Test GNN Examples and Add CUDA 12 Testing (#4317) @alexbarghi-nv
- Adds missing files to
update-version.sh
(#4314) @AyodeAwe - Fix stream synchronization in MTMG graph construction (#4275) @ChuckHastings
- Get rid of sentinel namespace (#4274) @PointKernel
- Update more C++ algorithm tests to include edge masked input graphs. (#4267) @seunghwak
- Update triangle count to internally use edge masking and update Triangle counting & K-hop neighbor tests to include edge masked graph view inputs (#4261) @seunghwak
- Add Additional Checks to
get_edgelist
andget_dask_edgelist
(#4256) @nv-rliu - Refactor Several MG Tests (#4244) @nv-rliu
v24.04.00
🚨 Breaking Changes
- C API for MNMG ECG (#4194) @naimnv
- Refactor and Fix
test_induced_subgraph_mg
(#4183) @nv-rliu - Removes APIs that have been deprecated and have exceeded the grace period, improves *_warning_wrapper performance (#4176) @rlratzel
- Fixes for pandas 2, latest cudf, and wheel building (#4144) @vyasr
🐛 Bug Fixes
- Update pre-commit-hooks to v0.0.3 (#4264) @KyleFromNVIDIA
- Removes (unintentional?) runtime dependency on
packaging
from nx-cugraph (#4240) @rlratzel - Fix kv_store_t build error when cuCollection is used as a backend and the value type is thrust::tuple (#4238) @seunghwak
- MTMG gather use of streams (#4237) @ChuckHastings
- remove NVTX calls in libcugraph_etl (#4229) @ChuckHastings
- Update cuGraph-PyG GraphSAGE Examples (#4224) @alexbarghi-nv
- transform_e bug fix in edge masking (#4221) @seunghwak
- Remove Usage of Experimental Datasets Package (#4220) @nv-rliu
- Fix pos_list api for force atlas (#4219) @hlinsen
- Fix
louvain_communities
,PropertyGraph
, cudfcolumn.full
, dgl 2.1.0 CI failures (#4215) @eriknw - Remove CUGRAPH_EXPECTS(!(this->has_edge_mask()), unimplemented.) (#4214) @naimnv
- fix scatter to handle empty graphs (#4193) @tingyu66
- Refactor and Fix
test_induced_subgraph_mg
(#4183) @nv-rliu - Update SG notebook (#4169) @naimnv
- Use input rng_state rather than creating a local one (#4168) @naimnv
- Fix
FutureWarnings
in Graph Creation (#4167) @nv-rliu - Fixes for pandas 2, latest cudf, and wheel building (#4144) @vyasr
- handle more RAPIDS version formats in update-version.sh, refactor dependencies.yaml (#4109) @jameslamb
- Mark kernels as internal (#4098) @robertmaynard
📖 Documentation
- Add Python 3.11 to Conda docs (#4213) @jakirkham
- Update installation docs for CUDA 12. (#4207) @bdice
- Fix in Docs to address question asked in issue 4182 (#4204) @acostadon
- [docs] fix GNN typo (#4196) @aMahanna
- Update README for cugraph notebooks (#4180) @naimnv
- Starting work on blog links and nx_cugraph docs (#4160) @acostadon
🚀 New Features
- Graph coloring and MIS (#4211) @naimnv
- Enable edge masking in the remaining primitives (#4186) @seunghwak
- Update per_v_random_select_transform_outgoing_e to support edge masking (#4174) @seunghwak
- cugraph examples (#4128) @naimnv
- Enable edge masking in additional primitives (#4126) @seunghwak
- Support CUDA 12.2 (#4088) @jameslamb
- cuGraph-DGL and WholeGraph Performance Testing with Feature Store Performance Improvements (#4081) @alexbarghi-nv
🛠️ Improvements
- Use
conda env create --yes
instead of--force
(#4286) @bdice - Always scatter in FP32 in TPConv (#4260) @tingyu66
- Remove outdated conda_build.sh. (#4257) @bdice
- update example build script with targets and options/help (#4254) @naimnv
- Refactor test utility functions and update SG & MG core_number tests to test with edge masking (#4253) @seunghwak
- nx-cugraph: add
is_negatively_weighted
(#4242) @eriknw - Add upper bound to prevent usage of NumPy 2 (#4233) @bdice
- Adds nx-cugraph benchmarks for APIs added to prior releases that were never benchmarked (#4228) @rlratzel
- Make external vertex and edge shuffling function public (#4227) @naimnv
- Updates nx-cugraph
README
for latest h/w, CUDA, python, NX requirements, moves updater to pre-commit (#4225) @rlratzel - Enable pytest failures on warnings from cudf (#4223) @mroeschke
- Added a weighted example to the jaccard notebook (#4222) @acostadon
- Use public pandas and cudf APIs where possible (#4218) @mroeschke
- Remove hard-coding of RAPIDS version where possible (#4217) @KyleFromNVIDIA
- Treat nx-cugraph, cugraph-pyg, cugraph-dgl, and cugraph-equivariant CI artifacts as pure wheels (#4216) @bdice
- Add degrees to C API (#4212) @ChuckHastings
- Add additional
kwargs
to GATConv (#4210) @tingyu66 - Add
DeprecationWarning
tobatched_ego_graphs
(#4209) @nv-rliu - Trim trailing whitespace (#4206) @bdice
- Automate C++ include file grouping and ordering using clang-format (#4205) @harrism
- nx-cugraph: add more shortest path algorithms (#4199) @eriknw
- Refactor MG Centrality Tests (#4197) @nv-rliu
- Add support for Python 3.11, require NumPy 1.23+ (#4195) @jameslamb
- C API for MNMG ECG (#4194) @naimnv
- nx-cugraph: update
get_info
function keys to NetworkX 3.3 (#4192) @eriknw - Adds additional detail to Jaccard notebook (#4189) @rlratzel
- Update devcontainers to CUDA Toolkit 12.2 (#4184) @trxcllnt
- Update
louvain_communities
to match NetworkX 3.3 (added max_level) (#4177) @eriknw - Removes APIs that have been deprecated and have exceeded the grace period, improves *_warning_wrapper performance (#4176) @rlratzel
- target branch-24.04 for GitHub Actions workflows (#4175) @jameslamb
- Add cuda-nvtx-dev, add missing CUDA library dependencies. (#4162) @bdice
- Enforce matching type (#4161) @jnke2016
- Add new all-pairs similarity algorithm (#4158) @ChuckHastings
- nx-cugraph: automatically generate trees in README.md (#4156) @eriknw
- Add environment-agnostic scripts for running ctests and pytests (#4152) @trxcllnt
- Ensure that
ctest
is called with--no-tests=error
. (#4148) @bdice - Update ops-bot.yaml (#4146) @AyodeAwe
- Replace local copyright check with pre-commit-hooks verify-copyright (#4130) @KyleFromNVIDIA
- Re-run
ci/release/update-version.sh 24.04.00
(#4127) @jakirkham - Address primitives performance regression with the added edge masking support (when edge masking is not in use) (#4119) @seunghwak
- Update cugraph for compatibility with the latest cuco (#4111) @PointKernel
- Add a new notebook for SNMG benchmark runs (#4091) @naimnv
- Ktruss implementation (#4059) @jnke2016
[NIGHTLY] v24.06.00
🔗 Links
🚨 Breaking Changes
- Call New
replicate_edgelist
Function (#4441) @nv-rliu - Enable FutureWarnings and DeprecationWarnings as errors in cugraph (#4415) @mroeschke
- Distributed Sampling in cuGraph-PyG (#4384) @alexbarghi-nv
- Move edge triangle count to the stable API (#4382) @jnke2016
- Address dask_cudf.read_csv chunksize deprecation (#4379) @mroeschke
- Convert device_memory_resource* to device_async_resource_ref (#4365) @harrism
- Update cugraph-pyg models for PyG 2.5 (#4335) @tingyu66
🐛 Bug Fixes
- Fix Broken
cpp_build
by Pinning Openmpi Build Ver (24.06) (#4475) @nv-rliu - [BUG] Pin Test Version of PyTorch to 2.1 to Resolve NCCL Error (#4464) @alexbarghi-nv
- Fix bug in kv_store_t's insertion methods (#4444) @naimnv
- Fix a bug in kv_store_t implementation (#4434) @naimnv
- [FIX] Remove Unintentional Expensive Check from SG Uniform Neighbor Sample (#4421) @alexbarghi-nv
- BFS fix for exec_policy (#4417) @ChuckHastings
- Skip/adjust doctests for deprecations (#4380) @mroeschke
- Address dask_cudf.read_csv chunksize deprecation (#4379) @mroeschke
- Fix bad Leiden result (#4376) @ChuckHastings
- Erdos-Renyi generator had bad logic in thrust calls (#4362) @ChuckHastings
- Change mamba option --force to --yes (#4357) @naimnv
- Fix false C++ HITS test failure on H100 (#4352) @seunghwak
- Bug fix in cugraph::detail::extract_transform_v_froniter_e (#4320) @seunghwak
- Refactor CMakeLists.txt to better express usage requirements (#4309) @robertmaynard
- [BUG] Always Persist Dask DataFrames in cuGraph-DGL Graph Storage (#4296) @alexbarghi-nv
- Fix Bug in MG
ego_graph
(#4262) @nv-rliu - [FIX] Remove Unnecessary CUDA Check Causing Deprecation Warning (#4239) @alexbarghi-nv
📖 Documentation
- Fixed links and added c++ docs per issue 4431 (#4435) @acostadon
- first copy of general cugraph tutorial. (#4396) @acostadon
- adding notebook to demo nx_cugraph (#4366) @acostadon
- Update operators.rst (#4339) @raybellwaves
- Removed obsolete methods from rst file (#4331) @acostadon
- Update DGL_support.md (#4327) @raybellwaves
- test sphinx mapping to networkx (#4323) @raybellwaves
- Updates the nx-cugraph README.md with the latest upstream URLs (#4321) @rlratzel
- Fix Table for Synthetic Data notebook (#4311) @nv-rliu
- Fix Docstring for
from_dask_cudf_edgelist
(#4310) @nv-rliu - fix PyPI URL in docs (#4282) @jameslamb
- DOC: doc-update-link-for-cugraphops (#4279) @raybellwaves
- Notebook fixes to formulas and outputs (#4272) @acostadon
🚀 New Features
- [FEA] New WholeGraph Feature Store for PyG (#4432) @alexbarghi-nv
- Biased sampling primitive (#4430) @seunghwak
- Distributed Sampling in cuGraph-PyG (#4384) @alexbarghi-nv
- [FEA] Support Seed Retention for Sampling with Renumbering (#4355) @alexbarghi-nv
- Always use a static gtest (#4344) @robertmaynard
- Update GNN sampling post processing functions to optionally consider seed vertices in renumbering (#4329) @seunghwak
- MNMG Approximation Algorithm for the Weighted Matching Problem (#4315) @naimnv
- [FEA] cuGraph GNN NCCL-only Setup and Distributed Sampling (#4278) @alexbarghi-nv
🛠️ Improvements
- Adds benchmark for
nx_cugraph.ego_graph
(#4451) @rlratzel - Pin torch version in
cugraph-dgl
wheel test (#4447) @tingyu66 - Call New
replicate_edgelist
Function (#4441) @nv-rliu - Fix MG Betweenness Centrality Test Bug (#4429) @nv-rliu
- Expose graph_view method to count multi edges through C API and PLC (#4426) @ChuckHastings
- Update doxygen comments to specify in,out instead of in/out (#4425) @ChuckHastings
- Enable FutureWarnings and DeprecationWarnings as errors in cugraph (#4415) @mroeschke
- Add MTMG support for Similarity and test for all-pairs Jaccard (#4411) @ChuckHastings
- Update collect_comm to handle value of tuple type (#4410) @naimnv
- Updates SG
PropertyGraph
andcugraph-service
to applyDataFrame.fillna()
based on latest cuDF changes (#4408) @rlratzel - Fix building cugraph with CCCL main (#4404) @trxcllnt
- Update decompress_to_edgelist to handle edge types (#4397) @naimnv
- nx-cugraph: add
ego_graph
(#4395) @eriknw - Remove unused benchmark fixture in test_dask_mg_random_walks (#4391) @mroeschke
- Address do_expensive_check FutureWarnings in Python tests (#4390) @mroeschke
- Address k_truss_subgraph(use_weights=) deprecation (#4389) @mroeschke
- Address PendingDeprecationWarning from _from_edgelist (#4388) @mroeschke
- Adjust deprecated cugraph.subgraph usage in Python tests (#4386) @mroeschke
- Update
k_truss
to work with many connected components (#4383) @eriknw - Move edge triangle count to the stable API (#4382) @jnke2016
- Replace deprecated cupy.sparse with cupyx.scipy.sparse (#4381) @mroeschke
- Migrate to
{{ stdlib("c") }}
(#4374) @hcho3 - Convert device_memory_resource* to device_async_resource_ref (#4365) @harrism
- add --rm and --name to devcontainer run args (#4361) @trxcllnt
- Update pip devcontainers to UCX v1.15.0 (#4360) @trxcllnt
- Add missing
<optional>
include (#4353) @trxcllnt nx-cugraph
: supportshould_run
that was added in NetworkX 3.3 (#4348) @eriknw- Update generation and flattening of dendrogram in Leiden (#4347) @ChuckHastings
- Update cugraph-pyg models for PyG 2.5 (#4335) @tingyu66
- Update pre-commit hooks to include all C and C++ files (#4332) @ChuckHastings
- Enable expression-based Dask Dataframe support (#4325) @rjzamora
- Use appropriate edge src and dst properties based on is_multi_gpu flag (#4319) @naimnv
- Test GNN Examples and Add CUDA 12 Testing (#4317) @alexbarghi-nv
- Adds missing files to
update-version.sh
(#4314) @AyodeAwe - Fix stream synchronization in MTMG graph construction (#4275) @ChuckHastings
- Get rid of sentinel namespace (#4274) @PointKernel
- Update more C++ algorithm tests to include edge masked input graphs. (#4267) @seunghwak
- Update triangle count to internally use edge masking and update Triangle counting & K-hop neighbor tests to include edge masked graph view inputs (#4261) @seunghwak
- Add Additional Checks to
get_edgelist
andget_dask_edgelist
(#4256) @nv-rliu - Refactor Several MG Tests (#4244) @nv-rliu
v24.02.00
🚨 Breaking Changes
- Remove Experimental Wrappers from GNN Code (#4070) @alexbarghi-nv
- Switch to scikit-build-core (#4053) @vyasr
- Update to CCCL 2.2.0. (#4052) @bdice
🐛 Bug Fixes
- Revert "Exclude tests from builds (#4147)" (#4157) @raydouglass
- Exclude tests from builds (#4147) @vyasr
- Constraint pytorch-dependent wheel test to only run on amd64 (#4133) @tingyu66
- Removes the
networkx_algorithm
decorator to all SCC functions to disable dispatching to them (#4120) @rlratzel - Correct
cugraph-pyg
package name used in wheels and fix test script (#4083) @tingyu66 - Fix Jaccard hang (#4080) @jnke2016
- Fix OOB error, BFS C API should validate that the source vertex is a valid vertex (#4077) @ChuckHastings
- [BUG]Fix non-type template parameter to cugraph::relabel (#4064) @naimnv
- Fix MG weighted similarity test failure (#4054) @seunghwak
- MG C-API test failure fixes (#4047) @seunghwak
- Add a barrier before cugraph Graph creation (#4046) @VibhuJawa
- Fix % 0 bug in MG_SELECT_RANDOM_VERTICES test (#4034) @seunghwak
- Branch 24.02 merge branch 23.12 (#4012) @vyasr
📖 Documentation
- Updates nx-cugraph README.md with latest algos (#4135) @rlratzel
- corrected links in C API and added groups for support functions (#4131) @acostadon
- Forward-merge branch-23.12 to branch-24.02 (#4049) @GPUtester
🚀 New Features
- Implement has_edge() & compute_multiplicity() (#4096) @seunghwak
- Update per_v_transform_reduce_incoming|outgoing_e to support edge masking (#4085) @seunghwak
- Remove Experimental Wrappers from GNN Code (#4070) @alexbarghi-nv
- MNMG ECG (#4030) @naimnv
- Replace graph_view.hpp::number_of_edges with compute_number_of_edges (#4026) @seunghwak
- Update count_if_e, transform_reduce_e, and transform_e to support edge masking (#4001) @seunghwak
- Sampling Performance Testing (#3584) @alexbarghi-nv
🛠️ Improvements
- Adds option to rapids_cpm_find for raft to disable hnswlib feature, adds updates for pytest 8 compat, temporarily skips IO intensive test in CI (#4121) @rlratzel
- Adds benchmarks for additional nx-cugraph 24.02 algos (#4112) @rlratzel
- nx-cugraph: use coverage to ensure all algorithms were run (#4108) @eriknw
- nx-cugraph: rename
plc=
to_plc=
(#4106) @eriknw - nx-cugraph: add
complement
andreverse
(#4103) @eriknw - nx-cugraph: add
core_number
(undirected graphs only) (#4100) @eriknw - nx-cugraph: add
is_tree
, etc. (#4097) @eriknw - Optimize the drop-duplicate functionality (#4095) @jnke2016
- nx-cugraph: add triangles and clustering algorithms (#4093) @eriknw
- nx-cugraph: PLC now handles isolated nodes; clean up our workarounds (#4092) @eriknw
- Remove usages of rapids-env-update (#4090) @KyleFromNVIDIA
- Provide explicit pool sizes and avoid RMM detail APIs (#4086) @harrism
- refactor CUDA versions in dependencies.yaml (#4084) @jameslamb
- build wheels for
cugraph-dgl
andcugraph-pyg
(#4075) @tingyu66 - Match weight-sharing option of GATConv in DGL (#4074) @tingyu66
- nx-cugraph: add weakly connected components (#4071) @eriknw
- nx-cugraph: indicate which plc algorithms are used and version_added (#4069) @eriknw
- Adds
nx-cugraph
benchmarks for 23.12 algos (SSSP, pagerank, hits, katz_centrality, degree_centrality, eigenvector_centrality) (#4065) @rlratzel nx-cugraph
: addto_undirected
method; add reciprocity algorithms (#4063) @eriknw- Switch to scikit-build-core (#4053) @vyasr
- Update to CCCL 2.2.0. (#4052) @bdice
- Prevent
actions/labeler
from addingLabel Checker
labels (#4048) @ajschmidt8 - Update dependencies.yaml to new pip index (#4045) @vyasr
- Remove checks for Pascal, no longer supported (#4044) @ChuckHastings
- Fix HITS convergence error. (#4043) @seunghwak
- Test select_random_vertices for all possible values of flags (#4042) @naimnv
- Remove CUGRAPH_BUILD_WHEELS and standardize Python builds (#4041) @vyasr
- Create
cugraph-equivariant
package (#4036) @tingyu66 - [FEA]: Add DASK edgelist and graph support to the Dataset API (#4035) @huiyuxie
- Add support for Louvain to MTMG (#4033) @ChuckHastings
- Clean up self-loop and multi-edge removal logic (#4032) @ChuckHastings
- Mtmg updates for rmm (#4031) @ChuckHastings
- nx-cugraph: adds
ancestors
,descendants
, and BFS algos (#4029) @eriknw - nx-cugraph: update usage of
nodes_or_number
for nx compat (#4028) @eriknw - Removes unsupported
setup.py
calls, cleans up text (#4024) @rlratzel - Resolves conflicts from forward-merging branch-23.12 into branch-24.02 (#4020) @rlratzel
- Add
HeteroGATConv
tocugraph-pyg
(#3914) @tingyu66 - Update for CCCL 2.x (#3862) @seunghwak
v23.12.00
🚨 Breaking Changes
- [BUG] Restore the original default order of CSR, which does not reverse edges in cuGraph-PyG (#3980) @alexbarghi-nv
Resultset
andDataset
Refactors (#3957) @nv-rliu- Moves more MG graph ETL to libcugraph and re-enables MG tests in CI (#3941) @jnke2016
🐛 Bug Fixes
- Pin actions/labeler to v4 (#4038) @raydouglass
- Find rmm before cuco (#4011) @vyasr
- Pin to minor versions of packages outside the cuGraph repository. (#4004) @bdice
- Move MTMG_TEST to MG tests block (#3993) @naimnv
- Fix Leiden refinement phase (#3990) @naimnv
- [BUG] Fix Graph Construction From Pandas in cuGraph-PyG (#3985) @alexbarghi-nv
- [BUG] Restore the original default order of CSR, which does not reverse edges in cuGraph-PyG (#3980) @alexbarghi-nv
- Fix eigenvector testing and HITS testing discrepancies (#3979) @ChuckHastings
- [BUG] Fix Incorrect Edge Index, Directory Selection in cuGraph-PyG Loader (#3978) @alexbarghi-nv
- [BUG] Check if Dask has quit to avoid throwing an exception and triggering a segfault on ddp exit (#3961) @alexbarghi-nv
- nx-cugraph: xfail test_louvain.py:test_threshold in Python 3.9 (#3944) @eriknw
📖 Documentation
- [DOC]: Fix invalid links and add materials to notebooks (#4002) @huiyuxie
- Update Broken Links in README.md (#3924) @nv-rliu
🚀 New Features
- Implement the transform_e primitive (to update property values for all edges) (#3917) @seunghwak
- Update the neighbor intersection primitive to support edge masking. (#3550) @seunghwak
🛠️ Improvements
- Correct defect found in DLFW testing (#4021) @ChuckHastings
nx-cugraph
README update: adds missingconnected_components
algo to table (#4019) @rlratzel- Build concurrency for nightly and merge triggers (#4009) @bdice
- Support
drop_last
Argument in cuGraph-PyG Loader (#3995) @alexbarghi-nv - Adds
update-version.sh
support for recently added files containing RAPIDS versions (#3994) @rlratzel - Use new
rapids-dask-dependency
metapackage for managingdask
versions (#3991) @galipremsagar - Fixes to nx-cugraph README: fixes typos, updates link to NX backend docs (#3989) @rlratzel
- Address FIXMEs (#3988) @seunghwak
- Updates README file to include nx-cugraph user documentation, adds nx-cugraph to main README (#3984) @rlratzel
- Update C API graph creation function signatures (#3982) @ChuckHastings
- [REVIEW]Optimize cugraph-DGL csc codepath (#3977) @VibhuJawa
- nx-cugraph: add SSSP (unweighted) (#3976) @eriknw
- CuGraph compatibility fixes (#3973) @brandon-b-miller
- Skip certain
cugraph-pyg
tests when torch-sparse is not available (#3970) @tingyu66 - nx-cugraph: add
eigenvector_centrality
,katz_centrality
,hits
,pagerank
(#3968) @eriknw - Cut peak memory footprint in graph creation (#3966) @seunghwak
- nx-cugraph: add CC for undirected graphs to fix k-truss (#3965) @eriknw
- Skip certain
cugraph-pyg
tests whentorch_sparse
is not available (#3962) @tingyu66 Resultset
andDataset
Refactors (#3957) @nv-rliu- Download
xml
docs artifact through CloudFront endpoint (#3955) @AyodeAwe - Add many graph generators to nx-cugraph (#3954) @eriknw
- Unpin
dask
anddistributed
for23.12
development (#3953) @galipremsagar - Errors compiling for DLFW on CUDA 12.3 (#3952) @ChuckHastings
- nx-cugraph: add k_truss and degree centralities (#3945) @eriknw
- nx-cugraph: handle seed argument in edge_betweenness_centrality (#3943) @eriknw
- Moves more MG graph ETL to libcugraph and re-enables MG tests in CI (#3941) @jnke2016
- Temporarily disable mg testing (#3940) @jnke2016
- adding C/C++ API docs (#3938) @BradReesWork
- Add multigraph support to nx-cugraph (#3934) @eriknw
- Setup Consistent Nightly Versions for Pip and Conda (#3933) @divyegala
- MTMG multi node (#3932) @ChuckHastings
- Use branch-23.12 workflows. (#3928) @bdice
- Fix an issue occurring in the cuGraph-DGL example for "mixed" mode. (#3927) @drivanov
- Updating Docs (#3923) @BradReesWork
- Forward-merge branch-23.10 to branch-23.12 (#3919) @nv-rliu
- new build all option (#3916) @BradReesWork
- Silence spurious compiler warnings (#3913) @seunghwak
- Link wholegrah and cugraphops XML docs (#3906) @AyodeAwe
- Updates to 23.12 (#3905) @raydouglass
- Forward-merge branch-23.10 to branch-23.12 (#3904) @GPUtester
- Build CUDA 12.0 ARM conda packages. (#3903) @bdice
- Merge branch-23.10 into branch-23.12 (#3898) @rlratzel
- Some MTMG code cleanup and small optimizations (#3894) @ChuckHastings
- Enable parallel mode (#3875) @jnke2016
- Adds benchmarks for
nx-cugraph
(#3854) @rlratzel - Add nx-cugraph notebook for showing accelerated networkX APIs (#3830) @betochimas
v23.10.00
🚨 Breaking Changes
- Rename
cugraph-nx
tonx-cugraph
(#3840) @eriknw - Remove legacy betweenness centrality (#3829) @jnke2016
- Remove Deprecated Sampling Options (#3816) @alexbarghi-nv
- cuGraph-PyG Loader Improvements (#3795) @alexbarghi-nv
- Expose threshold in louvain (#3792) @ChuckHastings
- Fix ValueError Caused By Batches With No Samples (#3789) @alexbarghi-nv
- Update to Cython 3.0.0 (#3716) @vyasr
🐛 Bug Fixes
- Add wget to test_notebook dependencies (#3918) @raydouglass
- Increase dask-related timeouts for CI testing (#3907) @jnke2016
- Remove
dask_cudf
dataframe for the_make_plc_graph
while creatingcugraph.Graph
(#3895) @VibhuJawa - Adds logic to handle isolated vertices at python layer (#3886) @naimnv
- Update Allocator Selection in cuGraph-DGL Example (#3877) @alexbarghi-nv
- Add file to update-version.sh (#3870) @raydouglass
- Fix torch seed in
cugraph-dgl
and-pyg
tests for conv layers (#3869) @tingyu66 - MFG C++ code bug fix (#3865) @seunghwak
- Fix subtle memory leak in nbr_intersection primitive (#3858) @ChuckHastings
- Uses
conda mambabuild
rather thanmamba mambabuild
(#3853) @rlratzel - Remove the assumption made on the client data's keys (#3835) @jnke2016
- Disable mg tests (#3833) @naimnv
- Refactor python code for similarity algos to use latest CAPI (#3828) @naimnv
- [BUG] Fix Batch Renumbering of Empty Batches (#3823) @alexbarghi-nv
- Temporarily disable the deletion of the dask dataframe (#3814) @jnke2016
- Fix OD shortest distance matrix computation test failures. (#3813) @seunghwak
- Use rapidsai/ci:cuda11.8.0-ubuntu22.04-py3.10 for docs build (#3811) @naimnv
- Fix ValueError Caused By Batches With No Samples (#3789) @alexbarghi-nv
- Update
python_run_cugraph
independencies.yaml
(#3781) @nv-rliu - Fixes
KeyError
forget_two_hop_neighbors
when called with a small start vertices list (#3778) @rlratzel
📖 Documentation
🚀 New Features
- WholeGraph Feature Store for cuGraph-PyG and cuGraph-DGL (#3874) @alexbarghi-nv
- similarity notebook to compare link prediction algos (#3868) @acostadon
- adding dining preference dataset (#3866) @acostadon
- Integrate C++ Renumbering and Compression (#3841) @alexbarghi-nv
- Sampling post processing functions to accelerate MFG creation. (#3815) @seunghwak
- [REVIEW] Add Pure DGL Dataloading benchmark (#3660) @VibhuJawa
🛠️ Improvements
- nx-cugraph: handle louvain with isolated nodes (#3897) @eriknw
- Pin
dask
anddistributed
for23.10
release (#3896) @galipremsagar - Updates the source build docs to include libcugraphops as a build prerequisite (#3893) @rlratzel
- fixes force atlas to allow string as vertex names (#3891) @acostadon
- Integrate renumbering and compression to
cugraph-dgl
to accelerate MFG creation (#3887) @tingyu66 - Enable weights for MG similarity algorithms (#3879) @jnke2016
- cuGraph-PyG MFG Creation and Conversion (#3873) @alexbarghi-nv
- Update image names (#3867) @AyodeAwe
- Update to clang 16.0.6. (#3859) @bdice
- Updates to build and test
nx-cugraph
wheel as part of CI and nightly workflows (#3852) @rlratzel - Update
cugraph-dgl
conv layers to use improved graph class (#3849) @tingyu66 - Add entry point to tell NetworkX about nx-cugraph without importing it. (#3848) @eriknw
- [IMP] Add ability to get batch size from the loader in cuGraph-PyG (#3846) @alexbarghi-nv
- Refactor legacy k truss (#3843) @jnke2016
- Use new
raft::compiled_static
targets (#3842) @divyegala - Rename
cugraph-nx
tonx-cugraph
(#3840) @eriknw - Add cuGraph devcontainers (#3838) @trxcllnt
- Enable temporarily disabled MG tests (#3837) @naimnv
- Remove legacy betweenness centrality (#3829) @jnke2016
- Use
copy-pr-bot
(#3827) @ajschmidt8 - Update README.md (#3826) @lmeyerov
- Adding metadata getter methods to datasets API (#3821) @nv-rliu
- Unpin
dask
anddistributed
for23.10
development (#3818) @galipremsagar - Remove Deprecated Sampling Options (#3816) @alexbarghi-nv
- [REVIEW] Cugraph dgl block improvements (#3810) @VibhuJawa
- Simplify wheel build scripts and allow alphas of RAPIDS dependencies (#3809) @vyasr
- Allow cugraph-nx to run networkx tests for nx versions 3.0, 3.1, and 3.2 (#3808) @eriknw
- Add
louvain_communities
to cugraph-nx (#3803) @eriknw - Adds missing copyright and license text to init.py package files (#3799) @rlratzel
- cuGraph-PyG Loader Improvements (#3795) @alexbarghi-nv
- Adds updates to build wheel and conda packages for
cugraph-nx
(#3793) @rlratzel - Expose threshold in louvain (#3792) @ChuckHastings
- Allow models to use a lightweight sparse structure (#3782) @tingyu66
- Clean-up old testing conventions in
test_ecg.py
(#3779) @nv-rliu - Calling
dataset.get_edgelist()
returns a copy of an edge list instead of global (#3777) @nv-rliu - Update dgl benchmarks (#3775) @VibhuJawa
- Forward-merge branch-23.08 to branch-23.10 (#3774) @nv-rliu
- Migrate upstream models to
cugraph-pyg
(#3763) @tingyu66 - Branch 23.10 merge 23.08 (#3743) @vyasr
- Update to Cython 3.0.0 (#3716) @vyasr
- Testing util improvements and refactoring (#3705) @betochimas
- Add new cugraph-nx package (networkx backend using pylibcugraph) (#3614) @eriknw
- New mtmg API for integration (#3521) @ChuckHastings
v23.08.00
🚨 Breaking Changes
- Change the renumber_sampled_edgelist function behavior. (#3762) @seunghwak
- PLC and Python Support for Sample-Side MFG Creation (#3734) @alexbarghi-nv
- Stop using setup.py in build.sh (#3704) @vyasr
- Refactor edge betweenness centrality (#3672) @jnke2016
- [FIX] Fix the hang in cuGraph Python Uniform Neighbor Sample, Add Logging to Bulk Sampler (#3669) @alexbarghi-nv
🐛 Bug Fixes
- Change the renumber_sampled_edgelist function behavior. (#3762) @seunghwak
- Fix bug discovered in Jaccard testing (#3758) @ChuckHastings
- fix inconsistent graph properties between the SG and the MG API (#3757) @jnke2016
- Fixes options for
--pydevelop
to remove unneeded CWD path ("."), restores use ofsetup.py
temporarily for develop builds (#3747) @rlratzel - Fix sampling call parameters if compiled with -DNO_CUGRAPH_OPS (#3729) @ChuckHastings
- Fix primitive bug discovered in MG edge betweenness centrality testing (#3723) @ChuckHastings
- Reorder dependencies.yaml channels (#3721) @raydouglass
- [BUG] Fix namesapce to default_hash and hash_functions (#3711) @naimnv
- [BUG] Fix Bulk Sampling Test Issue (#3701) @alexbarghi-nv
- Make
pylibcugraphops
optional imports incugraph-dgl
and-pyg
(#3693) @tingyu66 - [FIX] Rename
cugraph-ops
symbols (refactoring) and update GHA workflows to call pytest viapython -m pytest
(#3688) @naimnv - [FIX] Fix the hang in cuGraph Python Uniform Neighbor Sample, Add Logging to Bulk Sampler (#3669) @alexbarghi-nv
- force atlas notebook changes to run in cugraph 23.08 container. (#3656) @acostadon
📖 Documentation
- this fixes github links in cugraph, cugraph-dgl and cugraph-pyg (#3650) @acostadon
- Fix minor typo in README.md (#3636) @akasper
- Created landing spot for centrality and similarity algorithms (#3620) @acostadon
🚀 New Features
- Compute shortest distances between given sets of origins and destinations for large diameter graphs (#3741) @seunghwak
- Update primitive to compute weighted Jaccard, Sorensen and Overlap similarity (#3728) @naimnv
- Add CUDA 12.0 conda environment. (#3725) @bdice
- Renumber utility function for sampling output (#3707) @seunghwak
- Integrate C++ Sampling Source Behavior Updates (#3699) @alexbarghi-nv
- Adds
fail_on_nonconvergence
option topagerank
to provide pagerank results even on non-convergence (#3639) @rlratzel - Add Benchmark for Bulk Sampling (#3628) @alexbarghi-nv
- cugraph: Build CUDA 12 packages (#3456) @vyasr
🛠️ Improvements
- Pin
dask
anddistributed
for23.08
release (#3761) @galipremsagar - Fix
build.yaml
workflow (#3756) @ajschmidt8 - Support MFG creation on sampling gpus for cugraph dgl (#3742) @VibhuJawa
- PLC and Python Support for Sample-Side MFG Creation (#3734) @alexbarghi-nv
- Switch to new wheel building pipeline (#3731) @vyasr
- Remove RAFT specialization. (#3727) @bdice
- C API for renumbering the samples (#3724) @ChuckHastings
- Only run cugraph conda CI for CUDA 11. (#3713) @bdice
- Promote
Datasets
to stable and clean-up unit tests (#3712) @nv-rliu - [BUG] Unsupported graph for similiarity algos (#3710) @jnke2016
- Stop using setup.py in build.sh (#3704) @vyasr
- [WIP] Make edge ids optional (#3702) @VibhuJawa
- Use rapids-cmake testing to run tests in parallel (#3697) @robertmaynard
- Sampling modifications to support PyG and DGL options (#3696) @ChuckHastings
- Include cuCollection public header for hash functions (#3694) @seunghwak
- Refactor edge betweenness centrality (#3672) @jnke2016
- Refactor RMAT (#3662) @jnke2016
- [REVIEW] Optimize bulk sampling (#3661) @VibhuJawa
- Update to CMake 3.26.4 (#3648) @vyasr
- Optimize cugraph-dgl MFG creation (#3646) @VibhuJawa
- use rapids-upload-docs script (#3640) @AyodeAwe
- Fix dependency versions for
23.08
(#3638) @ajschmidt8 - Unpin
dask
anddistributed
for development (#3634) @galipremsagar - Remove documentation build scripts for Jenkins (#3627) @ajschmidt8
- Unpin scikit-build upper bound (#3609) @vyasr
- Implement C++ Edge Betweenness Centrality (#3602) @ChuckHastings