China’s agriculture and animal husbandry industry has entered a phase of rapid development. At the same time, the production process of agriculture and animal husbandry generates a large amount of greenhouse gas emissions. This thesis focuses on predicting carbon emissions from agriculture and animal husbandry, i.e., predicting carbon emissions in areas where cultivation and animal husbandry coexist. This thesis assumes that the future carbon emissions from agriculture and animal husbandry in Sichuan Province, China, are predictable. This thesis introduces the regression algorithm in machine learning the carbon emissions of agriculture and animal husbandry, mainly RF (Random Forest Regression Algorithm), and proposes that the combined model of carbon emissions of agriculture and animal husbandry is the FA-ACO-RF model (Factor Analysis-Ant Colony Optimization-Random Forest Regression Algorithm). Empirical evidence of carbon emission prediction in agriculture and animal husbandry was conducted in the Sichuan Province of China as an example, and the results showed that the prediction accuracy reached 99.19%. It is concluded that the FA-ACO-RF model can effectively and accurately predict carbon emissions from agriculture and animal husbandry in Sichuan Province.