Software

R Packages

R packages for spatial data processing, high-dimensional modeling, and binary classification diagnostics.

SpatPCA

Regularized principal component analysis for spatial data—identifies smooth, localized eigenfunctions and supports Kriging for new locations. Uses ADMM algorithm.

SpatMCA

Regularized maximum covariance analysis for spatial processes. Identifies smooth, localized patterns showing influence between processes. ADMM-based.

autoFRK

Thin-plate splines for basis function selection and nonstationary spatial covariance. Closed-form MLE, automatic selection via AIC. Handles large irregular datasets.

QuantRegGLasso

Adaptively weighted group Lasso for quantile regression. Variable selection and structure identification for varying-coefficient and additive models. Handles ultra-high dimensional covariates.

influenceAUC

Binary classification diagnostics—identifies influential observations and provides visualization tools for model performance and diagnostic insights.

Python

Python applications for MCP, conversational AI, computer vision, and deep learning.

Spherical DeepKriging

Spherical spatial prediction with DeepKriging and MRTS-sphere basis functions. PyPI package with CI, coverage, and Colab toy examples; code for DeepKriging on the Global Data.

KB Bridge

MCP server for intelligent knowledge base search. Context-aware retrieval with DSPy-powered query understanding, multi-stage search with content boosting, and answer quality reflection.

LLM Chatbot

Conversational AI framework with LangChain Agent. Integrates text-to-speech and speech-to-text from Hugging Face. Build chatbots with voice translation.