Code Carbon

With AI models becoming more ubiquitous and deployed across different sectors and industries, AI’s environmental impact is also growing. For this reason, it is important to estimate and curtail both the energy used and the emissions produced by training and deploying AI models.

Code Carbon enables developers to track carbon dioxide (CO₂) emissions across machine learning experiments or other programs.

For this purpose, it uses CO₂-equivalents [CO₂eq], which is a standardized measure used to express the global warming potential of various greenhouse gases: the amount of CO₂ that would have the equivalent global warming impact. For computing, which emits CO₂ via the electricity it is consuming, carbon emissions are measured in kilograms of CO₂-equivalent per kilowatt-hour. As a matter of fact, electricity is generated as part of the broader electrical grid by combusting fossil fuels for example.

Check out the Code Carbon website for more details!