Nvflare

What's New

Please check the [What's New](https://nvflare.readthedocs.io/en/2.4.0/whats_new.html) of document page.

What's Changed
* [2.4] Move workspace setup inside constructor by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2308
* [2.4] Update cli string 24 [skip ci] by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2309
* [2.4] Fix ParamsConverter signature by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2310
* Bump up the versions and improve dependency syntax by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2315
* [2.4] Improve monai integration tracking example by SYangster in https://github.com/NVIDIA/NVFlare/pull/2319
* [2.4] Exclude hidden Json file when parsing config by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2324
* [2.4] Update all examples/research/integration requirements by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2329
* [2.4] Removed the no need client custom path by yhwen in https://github.com/NVIDIA/NVFlare/pull/2321
* [2.4] update gnn readme by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2334
* [2.4] Fix CIFAR-10 global model selection by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2333
* [2.4] Enhance the handling of RC of task returned from clients by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2337
* [2.4] Address final VDR feedbacks by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2331
* [2.4] Cifar10 fix plot by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2345
* Added a few workarounds for HTTP driver's latency issues for 2.4 by nvidianz in https://github.com/NVIDIA/NVFlare/pull/2344
* [2.4] Address VDR report by SYangster in https://github.com/NVIDIA/NVFlare/pull/2336
* [2.4] Add notes about running TF with GPUs [skip ci] by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2347
* [2.4] Fix tb receiver by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2341


**Full Changelog**: https://github.com/NVIDIA/NVFlare/compare/2.4.0rc8...2.4.0rc9

What's Changed
* [2.4] Fix Auto login "communication error" by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2280
* [2.4] updates examples, add version matching note by SYangster in https://github.com/NVIDIA/NVFlare/pull/2281
* [2.4] Fixed the client_executor improper lock use. by yhwen in https://github.com/NVIDIA/NVFlare/pull/2288
* [2.4] vdr issues [skip ci] by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2286
* [2.4] Add notebook to GNN examples by ZiyueXu77 in https://github.com/NVIDIA/NVFlare/pull/2290
* [2.4] Fix FLModelUtil by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2292
* [2.4] Nemo examples fix by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2293
* [2.4] Check invalid input directory in nvflare config by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2294
* [2.4] Fix nemo peft notebook by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2296
* [2.4] Update nemo instructions by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2298
* [2.4] Update documentation and docstrings by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2300
* [2.4] Update execution api type documentation by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2304
* [2.4] address vdr feedback by SYangster in https://github.com/NVIDIA/NVFlare/pull/2299
* [2.4] Added debug headers for all message route in CoreCell for 2.4 Branch by nvidianz in https://github.com/NVIDIA/NVFlare/pull/2302
* [2.4] Rename to execution api type, update job command output by SYangster in https://github.com/NVIDIA/NVFlare/pull/2307


**Full Changelog**: https://github.com/NVIDIA/NVFlare/compare/2.4.0rc7...2.4.0rc8

What's Changed
* Fix the missing separators in nvflare cli by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2246
* Reduce CellPipe heartbeat error log by yanchengnv in https://github.com/NVIDIA/NVFlare/pull/2245
* Enhance the cloud script on the SSH inbound rule by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2249
* Fix p2p encrypt/decrypt by yanchengnv in https://github.com/NVIDIA/NVFlare/pull/2251
* Fix integration tests by SYangster in https://github.com/NVIDIA/NVFlare/pull/2254
* Fix POC HE mode, add HE step-by-step example by SYangster in https://github.com/NVIDIA/NVFlare/pull/2255
* Enhanced the client job status by yhwen in https://github.com/NVIDIA/NVFlare/pull/2247
* add Clara FL white paper [skip ci] by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2257
* added back the API auto login after server restart by yhwen in https://github.com/NVIDIA/NVFlare/pull/2256
* Fix CCWF with Client API, ParamsConverter supported_tasks by SYangster in https://github.com/NVIDIA/NVFlare/pull/2260
* Changed the TASK_UNKNOWN to log debug by yhwen in https://github.com/NVIDIA/NVFlare/pull/2262
* Fix simulator rc by yhwen in https://github.com/NVIDIA/NVFlare/pull/2263
* add notes for ``tree commands by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2264
* Add global round info for logging metrics global step by nvkevlu in https://github.com/NVIDIA/NVFlare/pull/2258
* Address few misc. bugs by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2252
* Fix launch_once=false connection error by SYangster in https://github.com/NVIDIA/NVFlare/pull/2269
* Fixed a case the admin has no auth_tokens for keycloak integration by yhwen in https://github.com/NVIDIA/NVFlare/pull/2268
* added handle for the empty return code file by yhwen in https://github.com/NVIDIA/NVFlare/pull/2261


**Full Changelog**: https://github.com/NVIDIA/NVFlare/compare/2.4.0rc6...2.4.0rc7

What's Changed
* Fix RestoreState callback by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2115
* Update gnn example by ZiyueXu77 in https://github.com/NVIDIA/NVFlare/pull/2114
* Update swarm learning example by SYangster in https://github.com/NVIDIA/NVFlare/pull/2064
* Merge duplicate sag_pt_deploy_map/sag_pt_ddp job templates [skip ci] by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2116
* Update the document on large model experiments. by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2118
* Fixes readthedocs config and link in docs [skip ci] by nvkevlu in https://github.com/NVIDIA/NVFlare/pull/2119
* NeMo peft example by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2094
* Changed default GPRC driver to sync version by nvidianz in https://github.com/NVIDIA/NVFlare/pull/2117
* Make client api pipe configurable by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2103
* Enhance TF Client API examples by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2056
* Fix docs build by SYangster in https://github.com/NVIDIA/NVFlare/pull/2123
* Print logs and write to logfile at the same time by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2120
* Job CLI Upgrade to support path notation by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2128
* Update configurations of LauncherExecutor by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2126
* Remove receive_global_model by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2129
* CLI Job upgrade, convert string to number if possible by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2130
* Update gnn template generation part by ZiyueXu77 in https://github.com/NVIDIA/NVFlare/pull/2131
* Fix gnn job templates link [skip ci] by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2135
* Updates for Client API examples, add cse and swarm by SYangster in https://github.com/NVIDIA/NVFlare/pull/2134
* Update NeMo PEFT example inference by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2124
* Fix doc format and docstring format [skip ci] by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2139
* Add discussion Q&A template [skip ci] by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2136
* enhance the rc handling for MPM. by yhwen in https://github.com/NVIDIA/NVFlare/pull/1985
* Update client api doc and docstrings by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2137
* Add NumPy + Client API example by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2031
* Remove mlflow.utils.time_utils dependency. by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2146
* Job Cli Upgrade: support custom configurations by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2143
* Job CLI Upgrade: including configuration key removal by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2147
* Add Step by step fed stats tabular example by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2151
* Update q-a.yml question rendering by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2154
* Support sys vars for job config and support parameterized template in job config by yanchengnv in https://github.com/NVIDIA/NVFlare/pull/2145
* Fix simulator by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2156
* Add release notes by SYangster in https://github.com/NVIDIA/NVFlare/pull/2144
* Added serialized size in the message header by nvidianz in https://github.com/NVIDIA/NVFlare/pull/2150
* Add TaskExchanger and FlareAgent by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2161
* Support graceful end_run processing by yanchengnv in https://github.com/NVIDIA/NVFlare/pull/2158
* Add Sklearn-linear step-to-step example by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2163
* Fix getting_started doc by SYangster in https://github.com/NVIDIA/NVFlare/pull/2169
* Add variable doc build version for github links by SYangster in https://github.com/NVIDIA/NVFlare/pull/2162
* Add experimental decorator and test by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2171
* Remove gen2 restriction from ad-hoc policy by yanchengnv in https://github.com/NVIDIA/NVFlare/pull/2170
* Fix Swarm learning README [skip ci] by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2172
* Fix job cli issues by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2141
* support getTask and submitResult timeout in job config by yanchengnv in https://github.com/NVIDIA/NVFlare/pull/2173
* Fixed the SystemVarName.SECURE_MODE error in simulator by yhwen in https://github.com/NVIDIA/NVFlare/pull/2174
* Experimental decorator for functions and classes by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2185
* Add example for HuggingFace SFT/PEFT by ZiyueXu77 in https://github.com/NVIDIA/NVFlare/pull/2092
* Add step-by-step tutorial for tabular data on svm, kmeans, and xgboost by ZiyueXu77 in https://github.com/NVIDIA/NVFlare/pull/2168
* Refactor LauncherExecutor and client api by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2164
* Update job templates and README by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2187
* Setup poc github rendering fix by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2188
* Add migration guide, add/update various sections by SYangster in https://github.com/NVIDIA/NVFlare/pull/2160
* Fix meta file processing in storage and improve schedule job retrieval by yanchengnv in https://github.com/NVIDIA/NVFlare/pull/2186
* update pub list for 2023 by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2178
* Fixed the missing simulator END_RUN event by yhwen in https://github.com/NVIDIA/NVFlare/pull/2182
* Enable run cross-validation without training workflow and examples by yhwen in https://github.com/NVIDIA/NVFlare/pull/2035
* Add overview, comm config, 3rd party integration docs [skip ci] by SYangster in https://github.com/NVIDIA/NVFlare/pull/2194
* Add metrics exchange mechanism by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2189
* change Fed Stats output format by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2199
* Simple FedAvg workflow by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2157
* Latest Werkzeug is not compatible with Flask by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2201
* Update links, example section, various cleanup [skip ci] by SYangster in https://github.com/NVIDIA/NVFlare/pull/2207
* Added PT Tensor decomposer registration on server side by nvidianz in https://github.com/NVIDIA/NVFlare/pull/2192
* KeyCloak authentication integration example by yhwen in https://github.com/NVIDIA/NVFlare/pull/2205
* Disable link check for example links by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2216
* POC bug fix by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2212
* Lightning api: add strict option for load_state_dict by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2211
* Check if fl model to be sent is empty by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2217
* save_config() function: fix to_hocon() issue by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2220
* Update xgboost example requirements [skip ci] by SYangster in https://github.com/NVIDIA/NVFlare/pull/2221
* Fixed a few issues with streaming error logging by nvidianz in https://github.com/NVIDIA/NVFlare/pull/2213
* fixed the custom_authentication example readme by yhwen in https://github.com/NVIDIA/NVFlare/pull/2218
* Add sag mlflow step by step example by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2208
* MONAI integration and metrics refactoring by SYangster in https://github.com/NVIDIA/NVFlare/pull/2223
* Don't run init() twice by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2224
* Fix LauncherExecutor racing issue by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2209
* misc doc enhancements, update monai metrics example readme by SYangster in https://github.com/NVIDIA/NVFlare/pull/2226
* Fix stats global key bug by SYangster in https://github.com/NVIDIA/NVFlare/pull/2229
* examples reformatting by SYangster in https://github.com/NVIDIA/NVFlare/pull/2228
* Fix hello examples README file and prepare data section by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2231
* Fix config read in lightning api by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2232
* POC Tutorial add HE support documentation in the notebook by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2230
* Add fl_ctx in convert method of ParamsConverter by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2233
* Display messages to users when cloud scripts unable to find public ip… by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2234
* Fix excepthook error by yanchengnv in https://github.com/NVIDIA/NVFlare/pull/2236
* Fix typo in model locator by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2239
* Update monai nvflare setup.py file by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2237
* az vm run-command invoke does not take multi-line scripts by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2214
* Fixed a race condition issue during the server start by yhwen in https://github.com/NVIDIA/NVFlare/pull/2235
* handle launcher failure condition in launch monitor by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2240
* Add he support to pt params converter by SYangster in https://github.com/NVIDIA/NVFlare/pull/2238


**Full Changelog**: https://github.com/NVIDIA/NVFlare/compare/2.4.0rc5...2.4.0rc6

What's Changed
* Adds support of multiple tasks to launcher executor by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2055
* Workaround for QA script by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2086
* Fix job cli boolean replacement issue by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2060
* Update ClientAPI PT examples and fix Lightning API by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2071
* Support cross site eval in FLModelUtil by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2089
* Fix for QA workflow scripts by SYangster in https://github.com/NVIDIA/NVFlare/pull/2090
* docs: fix typos [skip ci] by shuoer86 in https://github.com/NVIDIA/NVFlare/pull/2087
* Improve grpc 24 by yanchengnv in https://github.com/NVIDIA/NVFlare/pull/2088
* POC bug fixes: support multi-org, multi-roles project.yml by chesterxgchen in https://github.com/NVIDIA/NVFlare/pull/2049
* Fix cell and core cell by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2093
* Append new authorized user to blossom-ci whitelist [skip ci] by yinqingh in https://github.com/NVIDIA/NVFlare/pull/2096
* Fix typos in job templates by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2101
* Buffer List Support by nvidianz in https://github.com/NVIDIA/NVFlare/pull/2091
* Fixes for QA scripts by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2100
* Add docs sections and rearrangements [skip ci] by nvkevlu in https://github.com/NVIDIA/NVFlare/pull/2047
* Fix job cli create issue by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2095
* Last notebook example fix by IsaacYangSLA in https://github.com/NVIDIA/NVFlare/pull/2106
* Support both int and float by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2102
* Add server and client controlled cyclic step-by-steps by SYangster in https://github.com/NVIDIA/NVFlare/pull/2078
* Move RestoreOptimizers callback to lightning API by holgerroth in https://github.com/NVIDIA/NVFlare/pull/2107
* Major updates to GNN example by ZiyueXu77 in https://github.com/NVIDIA/NVFlare/pull/2109
* Fix site authorization by SYangster in https://github.com/NVIDIA/NVFlare/pull/2110
* Add None to is_primitive to support null in conf/json files by YuanTingHsieh in https://github.com/NVIDIA/NVFlare/pull/2111

New Contributors
* shuoer86 made their first contribution in https://github.com/NVIDIA/NVFlare/pull/2087
* yinqingh made their first contribution in https://github.com/NVIDIA/NVFlare/pull/2096

**Full Changelog**: https://github.com/NVIDIA/NVFlare/compare/2.4.0rc4...2.4.0rc5