v23.07.00
mdemoret-nv
released this
21 Jul 01:41
·
470 commits
to branch-23.11
since this release
🚨 Breaking Changes
- Use size_t for element counts & byte sizes (#1007) @dagardner-nv
- Fix logging of wrong feature_columns in CLI and exception in auto-complete (#983) @dagardner-nv
- Use new cudf C++ json writer (#888) @dagardner-nv
- Python 3.10 support (#887) @cwharris
- Update to CuDF 23.02 and MRC 23.07 (#848) @cwharris
🐛 Bug Fixes
- Fix fetch_example_data.py to work with s3fs v2303.6 (#1053) @dagardner-nv
- Fix models scripts/notebooks (#1051) @tzemicheal
- DFP visualization updates/fixes (#1043) @efajardo-nv
- Remove
loop
from severalasyncio
API calls (#1033) @mdemoret-nv - Fix release and dev container builds (#1027) @cwharris
- Adopt s3fs 2023.6 per #1022 (#1023) @dagardner-nv
- Move hard-coded path in E2E benchmarks to config json (#1011) @efajardo-nv
- Use size_t for element counts & byte sizes (#1007) @dagardner-nv
- Fix pybind11 link errors (#1003) @cwharris
- Fixing build with MRC breaking changes (#998) @mdemoret-nv
- Use cuda-toolkit in release container build (#997) @efajardo-nv
- Fix "No module named 'versioneer'" error (#990) @cwharris
- Warn and cast to cudf, when the C++ impl for MessageMeta receives a pandas DF (#985) @dagardner-nv
- Fix logging of wrong feature_columns in CLI and exception in auto-complete (#983) @dagardner-nv
- Remove duplicate log message in output when using pre-allocation (#981) @dagardner-nv
- Update to cuda-toolkit in release container build (#974) @efajardo-nv
- Add override for count method in SlicedMessageMeta (#972) @dagardner-nv
- Update .bashrc in container to activate morpheus conda environment (#969) @efajardo-nv
- Fix build issues & tests (#966) @dagardner-nv
- Remove/update tests which mock MRC builder objects (#955) @dagardner-nv
- Remove problematic test causing segfaults (#954) @dagardner-nv
- Fixing multi-sender stage configurations (#951) @mdemoret-nv
- Add support for fsspec.core.OpenFile instances to the MonitorStage (#942) @dagardner-nv
- Fix race condition in CompareDataFrameStage (#935) @dagardner-nv
- remove unnecessary quotes from CMAKE_ARGS expression (#930) @cwharris
- Enforce dtype for ColumnInfo and RenameColumn (#923) @dagardner-nv
- Fix early stopping code in dfencoder to use average loss of batches in validation set (#908) @hsin-c
- Fix
get_anomaly_score_losses
in dfencoder to work without categorical features (#893) @hsin-c - Update gnn_fraud_detection_pipeline & ransomware_detection examples to use the same version of dask (#891) @dagardner-nv
📖 Documentation
- Add note about needing docker-compose v1.28+ for DFP (#1054) @dagardner-nv
- Clean up CLX references in docstring (#1049) @efajardo-nv
- Update docs to use
docker compose
(#1040) @efajardo-nv - Fix documentation builds (#1039) @dagardner-nv
- GNN FSI modelcard++ (#1010) @tzemicheal
- DFP Model Card ++ (#1006) @hsin-c
- Document how to build the examples (#992) @dagardner-nv
- Documentation review edits (#911) @dagardner-nv
- Add PyTorch to ONNX section to training notebooks (#903) @efajardo-nv
- Update E2E Benchmarks README (#880) @efajardo-nv
🚀 New Features
- Update to CUDF 23.06 (#1020) @cwharris
- Grafana DFP dashboard (#989) @efajardo-nv
- DFP MultiFileSource optionally poll for file updates (#978) @dagardner-nv
- Migrate generic components from azure-ad workflow (#939) @bsuryadevara
- Migrate CLX parsers (#894) @efajardo-nv
- Python 3.10 support (#887) @cwharris
- GPUNetIO Integration (#879) @cwharris
- Update to CuDF 23.02 and MRC 23.07 (#848) @cwharris
🛠️ Improvements
- Support building the Morpheus containers when Morpheus is a submodule (#1057) @mdemoret-nv
- Changing GRPC requirement from
grpc-cpp
tolibgrpc
(#1056) @mdemoret-nv - Merge morpheus core spear phishing components. (#1044) @drobison00
- Example Documentation Updates (#1038) @cwharris
- Update docs and examples to use Triton 23.06 (#1037) @efajardo-nv
- SID visualization updates (#1035) @efajardo-nv
- New CI images with rapids 23.06 (#1030) @dagardner-nv
- Remove conda's base pip (CVE) (#1024) @pdmack
- Remove PIP_FIND_LINKS for torch from setup.py (#1019) @efajardo-nv
- Upgrade to pytorch 2.0.1 conda package (#1015) @efajardo-nv
- Model card ++ for ABP, Phishing, Root cause analysis (#1014) @gbatmaz
- Bring forward some CVE fixes from 23.03 release (#1002) @pdmack
- Remove patch from pybind11 (#1001) @dagardner-nv
- Install pytest-kafka via pip/pypi (#988) @dagardner-nv
- Adopt MatX 0.4.1 (#971) @dagardner-nv
- fix_contrib_instructions (#959) @yuchenz427
- Multi-class Sequence Classifier (#952) @efajardo-nv
- Add Pylint to CI (#950) @dagardner-nv
- Helper scripts for running CI locally (#949) @dagardner-nv
- Adopt updated utilities to pickup MatX 0.4.0 (#947) @dagardner-nv
- Add COLLABORATOR to exclusion in label-external-issues.yml (#946) @jarmak-nv
- Multiple Input Kafka Topics Support (#944) @bsuryadevara
- Tests for stages in DFP Production Example (#940) @dagardner-nv
- Integrate NVTabular into Morpheus Core and replace existing column_info based workflows. (#938) @drobison00
- Auto-Comment on External Issues (#926) @jarmak-nv
- DFP Integrated Training Updates (Stress Testing/Benchmarks) (#924) @bsuryadevara
- Add CONTRIBUTOR to triage label exception (#918) @jarmak-nv
- Improve DFP period functionality to allow for better sampling and ignoring period (#912) @mdemoret-nv
- Tests for gnn_fraud_detection_pipeline & ransomware_detection (#904) @dagardner-nv
- [ENH] Change external labeler to use the GH CLI for fine-grained token support (#899) @jarmak-nv
- [ENH] Label External Issues: Update secret, add discussions (#897) @jarmak-nv
- Use new cudf C++ json writer (#888) @dagardner-nv
- Create tests for examples with custom stages (#885) @dagardner-nv
- Use ARC V2 self-hosted runners for GPU jobs (#878) @jjacobelli
- Removing explicit driver install from CI runner (#877) @mdemoret-nv
- Adding an
update-version.sh
script and CI check to keep versions up to date (#875) @mdemoret-nv - Add DFP Viz dependencies to DFP container (#873) @efajardo-nv
- Use eval_batch_size for AutoEncoder loss stats (#861) @efajardo-nv
- GitHub Infra Update -Delete QST Issue Template in Favor of Discussions, Remove Add to Project Action (#860) @jarmak-nv
- Add dataset fixture to ease fetching DataFrames for tests (#847) @dagardner-nv
- Add DataManager object to morpheus/io (#846) @drobison00
- Suppress volatile warning when building code generated by Cython (#844) @dagardner-nv
- Replace usage of PreprocessLogParsingStage with PreprocessNLPStage (#842) @dagardner-nv
- Replace deprecated usage of make_node and make_node_full (#839) @dagardner-nv
- Bump version to 23.07 and version of MRC dep (#834) @dagardner-nv
- Update .gitignore to include some common temporary files that shouldn't be tracked by git (#829) @dagardner-nv
- Pre-allocate needed columns in abp_pcap_detection example (#820) @dagardner-nv
- Use ARC V2 self-hosted runners for CPU jobs (#806) @jjacobelli