Skip to content
#

woe-and-iv

Here are 17 public repositories matching this topic...

CredVibe

CredVibe is an ML credit scorecard system achieving 95%+ default recall with explainable predictions for loan risk assessment. Features KS/Gini validation, Optuna tuning, FastAPI + Streamlit deployment. Generates CIBIL-like scores, and converts to business rules for BRE integration.

  • Updated Feb 19, 2026
  • Python

End-to-end credit risk system on 1.3M loans — WoE/XGBoost PD model (AUROC 0.767), local LLM signal extraction (99.9% parse rate), SHAP-grounded credit memo generator, and a multi-modal Early Warning System combining FinBERT news sentiment with FRED macro signals (recall 1.00).

  • Updated Mar 15, 2026
  • Jupyter Notebook

Improve this page

Add a description, image, and links to the woe-and-iv topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the woe-and-iv topic, visit your repo's landing page and select "manage topics."

Learn more