13 projects
torchpsort
Fast, differentiable sorting in pure PyTorch without C++ or CUDA
mcqgam
Monotone Composite Quantile Generalized Additive Model
xuplift
Explainable uplift modeling via linearized kernel feature maps.
striders
Efficient surrogate-based model explanations (XAI) using landmark-based kernel approximations for scalable SHAP values.
mqboost
Monotonic composite quantile gradient boost regressor
acachetools
Async cachetools decorators with stampede protection.
colablinter
Linting and formatting Python code in Google Colab.
fastbinning
A high-performance binning library specifically designed for Credit Risk Modeling and Scorecard Development.
rektrag
No-brainer vectorless RAG combining docling and toon-python
imlightgbm
LightGBM for label-imbalanced data with focal and weighted loss function
autolgb
None
rektgbm
No-brainer machine learning solution to achieve satisfactory performance
quantile-tree
Monotone quantile regressor
