Add support for diabetes dataset - #58

Open

faildeny wants to merge 29 commits into

mainfrom

add_diabtest_dataset

Commits on Nov 13, 2025

Add support for Diabetes dataset
faildeny
committed
08b18f0
Add Balanced Random Forest selection as separate model
faildeny
committed
4d0bc62
Minor xgb fix for results compatibility
faildeny
committed
5aefbd6
Update tests with Diabetes dataset and unlock models testing
faildeny
committed
4da5e2e
Fix for config independent tests
faildeny
committed
891c7d7
Update github actions to meet latest github system changes
faildeny
committed
df5d480

Commits on Feb 3, 2026

Add AUROC calculation and switch to predict_proba in models
faildeny
committed
3e3a0ef
Improve Random Forest hyperparameters
faildeny
committed
fa7191d
Minor changes with report generation
faildeny
committed
157fd8c
Move to one commonm federated preprocessing function for public datasets.
faildeny
committed
185789a
Add scripts for automated benchmarking
faildeny
committed
af74c35
Add notebook for results visualisation.
faildeny
committed
5094455

Commits on Feb 4, 2026

Commits on Feb 5, 2026

Add minimum num of samples in dirichlet partitioning
faildeny
committed
5099403
Add threshold fine tuning on validation set for all models
faildeny
committed
9005e97
Fix missing metrics from XGBoost model in distributed fit
faildeny
committed
67f20da

Commits on Feb 9, 2026

Fix xgb training with device argument
faildeny
committed
183ab75
Add usage of seed from config for reproducibility
faildeny
committed
d594349

Commits on Feb 10, 2026

Update benchmarking parameters
faildeny
committed
3aca054
Rename old XGBoost implementation to xgblr
faildeny
committed
ad77488
Rename old XGBoost implementation to xgblr tests
faildeny
committed
90a42cd
Update gitignore
faildeny
committed
5e80ec5

Commits on Feb 23, 2026

Commits on Apr 2, 2026

Add fairness metrics computation
faildeny
committed
32eb2b7