PAMLA: E2E Prototype Exhibition of a Predictive Ag ML Assistant

https://github.com/devinkristopher/PAMLA PAMLA (Predictive Agriculture Machine Learning Assistant) is a locally-hosted, open-source agricultural forecasting AI assistant for machine learning-based predictive agriculture. PAMLA uses hyperlocal environmental telemetry collected from local outdoor sensor networks to forecast soil moisture on a 24-hour prediction horizon. While PAMLA's data preprocessing pipeline is designed to integrate with the home automation platform Home Assistant (as it expects Home Assistant-exported CSV telemetry), the general forecasting and inference architecture is platform-agnostic once data is transformed into PAMLA’s expected schema. Because the telemetry expected by the data preprocessing pipeline is a CSV file of commonplace environmental metrics, PAMLA's deployment contract could be reasonably fulfilled by non-supported systems that provide equivalent environmental data with minor data preparation.