eco2AI: Carbon Emissions Tracking of Machine Learning Models as the First Step Towards Sustainable AI

Abstract— The size and complexity of deep neural networks used in AI applications continue to grow exponentially, significantly increasing energy consumption for training inference by these models. We introduce an open-source package eco2AI help data scientists researchers track the equivalent CO 2 emissions their models a straightforward way. In we focus on accurate tracking regional accounting. encourage research community search new optimal Artificial Intelligence (AI) architectures with lower computational cost. motivation also comes from concept AI-based greenhouse gases sequestrating cycle both Sustainable Green pathways. code documentation are hosted Github under Apache 2.0 license https://github.com/sb-ai-lab/Eco2AI .