Yashwanthsairam/infra-predictive-optimization-mlops-lite

Predictive Resource Optimization — Decision Dashboard (Gradio-Lite)

Runs entirely in your browser via Pyodide -- no server. Forecasts CPU and memory utilization 5-30 minutes ahead for on-premises virtualized data center VMs, using a Random Forest model trained on the Bitbrains GWA-T-12 FastStorage telemetry trace (QM640 capstone). Model files are fetched live from the Hugging Face Model Hub and inference runs client-side.