A game theoretic approach to explain the output of any machine learning model.

Efficient surrogate-based model explanations (XAI) using landmark-based kernel approximations for scalable SHAP values.

Linting and formatting Python code in Google Colab

High-performance binning library specifically designed for Credit Risk Modeling and Scorecard Development.

No-brainer vectorless RAG combining docling and toon-python

[JCGS 2021] Official Implement of the paper "Learning Multiple Quantiles With Neural Networks"

LightGBM for handling label-imbalanced data with focal and weighted loss functions in binary and multiclass classification

No-brainer model combining LightGBM with hyperparameter tuning using Optuna

Explainable Boosted Scoring with Python: turning XGBoost, LightGBM, and CatBoost into explainable scorecards

No-brainer model combining LightGBM and XGBoost with hyperparameter tuning using Optuna

Monotone composite quantile regression neural network (MCQRNN) with tensorflow 2.x

Multiple quantiles estimation model maintaining non-crossing condition (or monotone quantile condition) and preventing quantile crossing using LightGBM and XGBoost