Additive manufacturing of light metal alloys has gained momentum lately in both academic and industrial domains. This necessitates a much deeper analysis of such materials under 3D printing processes to fully understand their behaviors corresponding to different sets of process parameters and predict the mechanical properties of the final fabricated items. Furthermore, the promising results accomplished by machine learning (ML) approaches in different fields of study are believed to guarantee reliability of these methods in forecasting the characteristics of the additively manufactured light metal alloys as well. In this paper, a comprehensive dataset regarding the melt pool morphology of stainless steels as the most common light metal alloys studied in the literature under the widely recognized selective laser melting process has been developed to train a wide variety of ML models. These models not only demonstrated exceptional performances in comparison with previous efforts and introduced new state-of-the-art melt pool geometry approximators but they also acknowledged the validity of the data set developed.