Accuracy is not the only Metric that Matters: Estimating Energy Consumption fo Deep Learning Models

This page is about our paper

Accuracy is not the only Metric that Matters: Estimating Energy Consumption fo Deep Learning Models
by Johannes Getzner, Bertrand Charpentier, and Stephan Günnemann
Published at the Tackling Climate Change with Machine Learning  Workshop (Climate Change AI - ICLR)  2023. Spotlight talk.

Abstract

Modern machine learning models have started to consume incredible amounts of energy, thus incurring large carbon footprints (Strubell et al., 2019). To address this issue, we have created an energy estimation pipeline1, which allows practitioners to estimate the energy needs of their models in advance, without actually running or training them. We accomplished this, by collecting high-quality energy data and building a first baseline model, capable of predicting the energy consumption of DL models by accumulating their estimated layer-wise energies

Links

[Paper|Video|Github]