logreg

Multi-model local AI agent demo — five small models (gemma3, qwen3.5, embeddinggemma, GLM-OCR, whisper/piper) collaborating on intent, multi-steps, tools, RAG, and voice.

This repository uses CHIRPS satellite and local station blended precipitation data to analyze and classify drought risk in California’s Central Valley. The focus is to prepare a drought classifier, visualization & analysis, and reproducible ML pipeline.

Today we will help you with basic tasks involved in machine learning in Python.

For Macroecology class - a simple GLM approach to predict tree specie distribution based on environmental predictors.