How to estimate carbon footprint when training deep learning models? A guide and review
This provides a practical guide for AI practitioners to assess environmental impact, but it is incremental as it synthesizes existing tools without introducing new methods.
The paper tackles the problem of estimating the carbon footprint of training deep learning models by reviewing and comparing existing tools, finding that tool estimates vary significantly across different neural networks and server types.
Machine learning and deep learning models have become essential in the recent fast development of artificial intelligence in many sectors of the society. It is now widely acknowledge that the development of these models has an environmental cost that has been analyzed in many studies. Several online and software tools have been developed to track energy consumption while training machine learning models. In this paper, we propose a comprehensive introduction and comparison of these tools for AI practitioners wishing to start estimating the environmental impact of their work. We review the specific vocabulary, the technical requirements for each tool. We compare the energy consumption estimated by each tool on two deep neural networks for image processing and on different types of servers. From these experiments, we provide some advice for better choosing the right tool and infrastructure.