Feature enhancements

What's Changed

• Change BaseException to Exception by @YuanTingHsieh in #1790
• Allow alternative Configurations: Part 3. Config Service Integration by @chesterxgchen in #1824
• Include exception type in error log by @YuanTingHsieh in #1839
• Fix doc [skip ci] by @YuanTingHsieh in #1841
• Add Launcher spec and SubprocessLauncher by @YuanTingHsieh in #1763
• Add LauncherExecutor by @YuanTingHsieh in #1783
• Adds ModelExchanger by @YuanTingHsieh in #1837
• Fix auth integration test by @YuanTingHsieh in #1842
• Fix ACK Timeout Issue [skip ci] by @nvidianz in #1836
• Fixed a concurrency issue by @nvidianz in #1848
• New Cell to provide layers to support original Cell API by @IsaacYangSLA in #1832
• Revert "Remove print/logger" by @IsaacYangSLA in #1853
• Fix SNI passing for AdminAPI (#1849) by @xander-aphe-hatschi in #1850
• FedOpt with batchnorm by @holgerroth in #1851
• Fix auto integration tests by @YuanTingHsieh in #1847
• Add unit tests of SubprocessLauncher and ModelExchanger by @YuanTingHsieh in #1845
• Add Weights and Biases for experiment tracking by @nvkevlu in #1838
• New cell fix by @IsaacYangSLA in #1861
• Fixes an End-of-Stream bug by @nvidianz in #1862
• remove abort_task command by @nvkevlu in #1854
• Update documentation for experiment tracking, peer context [skip ci] by @nvkevlu in #1859
• Add NVFlare Client API by @YuanTingHsieh in #1863
• Add client_api example by @YuanTingHsieh in #1844
• Add decorator example by @YuanTingHsieh in #1855
• Add lightning API and example by @YuanTingHsieh in #1867
• Refactor Learner api by @yanchengnv in #1761
• Separate plot-requirements out by @YuanTingHsieh in #1869
• Expose FLModel and change to send/receive by @YuanTingHsieh in #1874
• Updates ML-to-FL Client API by @YuanTingHsieh in #1877
• Fixes the streaming problem in simulator by @nvidianz in #1880
• Enable NewCell by @IsaacYangSLA in #1882
• Refactor xgboost examples by @SYangster in #1876
• Update docs, tutorials, examples, and integration tests with POC updates by @nvkevlu in #1873
• Fix examples and FLModel by @YuanTingHsieh in #1883
• Fix integration test by @YuanTingHsieh in #1884
• ML2FL examples: small changes by @chesterxgchen in #1891
• Add garbage collection at the end of each round in cyclic controller by @IsaacYangSLA in #1890
• Fix integration test configs timing issues by @YuanTingHsieh in #1886
• add Mac specific setup by @chesterxgchen in #1892
• update config and config factory by @chesterxgchen in #1893
• remove wget in df_stats example by @chesterxgchen in #1878
• Fixed net agent to work with NewCell by @yanchengnv in #1897
• Cherry-picked from 2.3.2 #1868 by @IsaacYangSLA in #1895
• Revert some changes from POC update commit that caused issues (#1894) by @YuanTingHsieh in #1904
• Enhance lightning example by @YuanTingHsieh in #1903
• Improve NewCell send request timeout handling by @IsaacYangSLA in #1899
• Add metrics exchanger and retriever by @YuanTingHsieh in #1889
• Support admin client handlers and submit_job with custom data by @yanchengnv in #1881
• Cherry-pick Add client heartbeat log (#1864) by @YuanTingHsieh in #1901
• Update references to dev after change in branch, update broken links [skip ci] by @nvkevlu in #1908
• Fix MonaiAlgo args [skip ci] by @holgerroth in #1914
• Fixed the admin tool requires the resources path. by @yhwen in #1910
• Fix FLAdminAPI by @YuanTingHsieh in #1916
• Use future waiter by @IsaacYangSLA in #1909
• Changed the simulator to use client_runner.run_abort_signal.triggered by @yhwen in #1918
• Fix typos [skip ci] by @YuanTingHsieh in #1915
• Make sure the Simulator supports alternative config formats by @chesterxgchen in #1923
• Enhance lightning api, client api, LauncherExecutor by @YuanTingHsieh in #1919
• Add reasearch folder for "ConDistFL: Conditional Distillation for Federated Learning from Partially Annotated Data" [skip ci] by @holgerroth in #1929
• Update urls by @holgerroth in #1933
• Wrap training scripts inside main method by @YuanTingHsieh in #1939
• Fixed the recursive FLComponents creation. by @yhwen in #1934
• NewCell and Cell renaming by @IsaacYangSLA in #1941
• Add experiment tracking with MONAI for MetricExchanger by @nvkevlu in #1925
• Re-add cli persistent history by @SYangster in #1938
• Job CLI: create Job, submit Job, list_templates, show_variables by @chesterxgchen in #1888
• POC Upgrade 2 by @chesterxgchen in #1944
• CC SDK 1.0 integration by @IsaacYangSLA in #1926
• Rewrite double quote to heredocs to avoid bash/zsh issues by @IsaacYangSLA in #1947
• Fix controller unit test timing by @YuanTingHsieh in #1937
• make sure the Job CLI support multi-config formats by @chesterxgchen in #1946
• add poc workspace to the nvflare config command by @chesterxgchen in #1948
• update POC tutorials by @chesterxgchen in #1949
• update docs after change in nvflare poc command [skip ci] by @nvkevlu in #1945
• Add Sean to build command by @YuanTingHsieh in #1950
• Vertical XGBoost with PSI integration by @SYangster in #1922
• Allow metric negation in model selection by @holgerroth in #1951
• add deprecation commands by @chesterxgchen in #1952
• Secure message by @IsaacYangSLA in #1954
• Fix fl model utils by @YuanTingHsieh in #1902
• SFM Heartbeat Support by @nvidianz in #1942
• Enhance ML2FL API by @YuanTingHsieh in #1953
• Remove fobs calls by @yhwen in #1960
• Removed the not used import in cell.py (temporary) by @yhwen in #1961
• Improved error handling and fixed memory leak by @nvidianz in #1921
• Fix unit test and integration tests by @YuanTingHsieh in #1962
• Client controller by @yhwen in #1913
• Update POC tutorials and fix POC bugs by @chesterxgchen in #1958
• Add Job CLI Tutorials and step-by-step initial examples by @chesterxgchen in #1957
• Refactor format_log_message with more readability by @IsaacYangSLA in #1965
• Remove some POC stop message by @chesterxgchen in #1966
• Add experiment tracking docs by @nvkevlu in #1963
• Add SimpleCellCipher to replace session key manager and cell cipher by @IsaacYangSLA in #1964
• Improve CLI command error handling by @chesterxgchen in #1971
• Polish notebook for Job CLI [skip ci] by @nvkevlu in #1975
• Add Step by step example part 1: CIFAR10 Image Stats by @chesterxgchen in #1972
• Add implementations to research folder for "ConDistFL: Conditional Distillation for Federated Learning from Partially Annotated Data" [skip ci] by @nanaHa1003 in #1940
• Improve ML2FL example documentation by @YuanTingHsieh in #1969
• Step by step Example CIFAR10 part 2 : SAG Example by @chesterxgchen in #1978
• Job CLI bug fix: make sure we use the existing meta.conf from job_template by @chesterxgchen in #1980
• Optimize workspace saving and download_job command changes by @yhwen in #1979
• Add SFT example to NeMo integration by @ZiyueXu77 in #1774
• Update publication list by @holgerroth in #1976
• Client Controlled Workflow by @yanchengnv in #1967
• prepare the download_job data files. by @yhwen in #1984
• Secure Messaging by @nvidianz in #1983
• Add TensorFlow client api example by @YuanTingHsieh in #1982
• Datum Support in FOBS by @nvidianz in #1986
• Support big job workspace by @yanchengnv in #1988
• Add broadcast and aux_communication channel by @IsaacYangSLA in #1991
• Add secure argument to Task and set secure argument in cell's call. by @IsaacYangSLA in #1992
• Kill stream threads on shutdown by @nvidianz in #1995
• Fix a multithread race condition in broadcast of Cell by @IsaacYangSLA in #1998
• Added GNN example and updated readme by @wangxiaoyunNV in #1968
• Add a feature to run dashboard locally without docker by @IsaacYangSLA in #2004
• Enhance Job CLI to support multi-apps/sites by @chesterxgchen in #2001
• Support secure task in CCWF and comm improvement by @yanchengnv in #1999
• Add fl meta in lightning api by @YuanTingHsieh in #1989
• Add benchmark scores to SFT model evaluation by @ZiyueXu77 in #1987
• Update vertical xgb examples for gpu support by @SYangster in #1977
• Add additional section to Job cli tutorial by @chesterxgchen in #2006
• Fix GNN example graph [skip ci] by @chesterxgchen in #2010
• add few more licenses [skip ci] by @chesterxgchen in #2011
• Fix license header years [skip ci] by @YuanTingHsieh in #1996
• Add step-by-step example for CIFAR10 site-specific configs and deploy_map by @chesterxgchen in #2009
• Removed default keepalive parameters from GRPC driver by @nvidianz in #2012

New Contributors

• @xander-aphe-hatschi made their first contribution in #1850
• @nanaHa1003 made their first contribution in #1940
• @wangxiaoyunNV made their first contribution in #1968

Full Changelog: 2.4.0rc1...2.4.0rc2