Energy consumption is a major factor to plan for when implementing a long-term project or service that uses large-scale machine learning algorithms. Now, a team of researchers from Georgia Tech has created an interactive tool called EnergyVis that allows users to compare energy consumption across locations and against other models. "Sometimes, training machine learning models from end-to-end takes the same amount of energy as a transatlantic flight. Is every organization using machine learning able to budget for such an expense? What if the grid in which a business runs is running on coal versus green energy?" said Omar Shaikh, a computer science undergraduate student.

The advent of larger machine learning (ML) models have improved state-of-the-art (SOTA) performance in various modeling tasks, ranging from computer vision to natural language. As ML models continue increasing in size, so does their respective energy consumption and computational requirements. However, the methods for tracking, reporting, and comparing energy consumption remain limited. We presentEnergyVis, an interactive energy consumption tracker for ML models. Consisting of multiple coordinated views, EnergyVis enables researchers to interactively track, visualize and compare model energy consumption across key energy consumption and carbon footprint metrics (kWh and CO2), helping users explore alternative deployment locations and hardware that may reduce carbon footprints. EnergyVis aims to raise awareness concerning computational sustainability by interactively highlighting excessive energy usage during model training; and by providing alternative training options to reduce energy usage.