Secretflow

Added

- [component] Add stats psi component
- [component] Add score card component
- [component] Add data sampling component
- [component] Add type cast component
- [component] Optimization GLM & LR training speed by using beaver cache
- [SLModel] Add attack method BLA
- [SLModel] Add defense method CAE
- [sgb] Support sample weight training

Fix

- [component] The online prediction of the GLM model has a large deviation compared with the offline prediction.

Fix

- [component] fix sgb export model sigmoid type not consistent with the offline prediction sigmoid type
- [component] fix sgb/xgb export model miss base score

Added

- [component] pvalue support GLM
- [component] onehot add drop parameter to support first & mode
- [component] XGB/SGB support checkpoint
- [component] LR support report weights
- [component] Add union component
- [FLModel/SLModel] tf backend support custom loss
- [SLModel] Base model support additional loss
- [SLModel] Add attack method FSHA
- [SLModel] Add defense method MID
- [SLModel] Add defense method FedPass
- [Infra] Add arm64 build support

Added

- SL: custom trainning step supported by lightning style base module.

Removed

- No longer providing x64 macOS binary packages.

Added

- SL: grad_average is supported on GPU.
- SL: mix_up defence can be called through callback.
- Simulation: add some unpartitioned datasets.

Fixed

- Fix order in componet load table.

Changed

- Serving: Update serving linkfunc type.
- FL: rename get/set_gradients methods
- SGX: improve doc for the accuracy analysis.
- SL: import fia attack.

Added

- Add gradient_average callback in SL.

Fixed

- Fix tuner default resource allocate.

Changed

- Boost the speed of autoattack benchmark.
- Upgrade to Python 3.10.