model-diagnostics

PiML (Python Interpretable Machine Learning) toolbox for model development & diagnostics

Implementation of NAACL 2024 Outstanding Paper "LM-Infinite: Simple On-the-Fly Length Generalization for Large Language Models"

🔍 AI-powered diagnosis for Scikit-learn models: Detect overfitting, data leakage, class imbalance & more with LLM-generated insights

Laplace's Scissors: protected readout rotations for open LLMs. Diagnose weak output directions, repair answers, preserve successful generations, and analyze scaling deviations with Rust and Triton on CUDA/HIP.

This repository commits to the application of biostatistics knowledge on clinical, randomized trials and observational studies.

Tool for evaluating atmospheric carbon dioxide concentrations as simulated by Earth system models

Sub-package of spatstat containing functionality for parametric modelling and inference

Polyhedron is a domain-driven Python optimization modeling framework that turns business entities into transparent MILP/MIQP models, with built-in diagnostics, scenario analysis, and robust solver interoperability.

Auto-diagnose deep learning training pathologies: overfitting, unstable LR, exploding gradients, noisy training, and early stopping issues.

This project uses the Reaction Time Survey dataset to develop a linear regression model for accurately predicting student reaction times based on various predictors. Tech: R (RStudio)

Tool for evaluating atmospheric carbon dioxide concentrations as simulated by Earth system models

This repository contains some of the time series analysis, diagnostics and forecasting projects I have done.

Global challenge to create Species Distribution Model to predict occurrence of frog species, Litoria fallax, in Australia.

Nine diagnostic tools for detecting and understanding overfitting in scikit-learn models — polynomial overfitting, learning curves, validation curves, bias-variance decomposition, regularisation sweeps, data leakage detection, and more. Companion code for the ML Diagnostics Mastery series.

Low-cost plasticity diagnostics for reinforcement learning models.

Explain why your model fails — not just how accurate it is.

Approximation Bayesian Computation: Population Monte Carlo in MATLAB and Python

time series analysis in R use cases

One-call diagnostics for statistical models in Python, a port of R's easystats/performance

🩺 Diagnose model issues in scikit-learn with evidence-based insights and LLM-powered analysis for improved machine learning performance.