Abstract: The I-beam composite stiffened panel is a common engineering structure in aircraft load-bearing components. The axial compressive buckling performance of such panels is typically investigated using engineering methods and finite element analysis. However, these conventional approaches are characterized by low accuracy and high computational time, making it difficult to achieve efficient and precise research. In this study, an efficient machine learning framework is established to address the prediction of the buckling load and buckling mode shape of I-beam composite stiffened panels under axial compression. By designing the sample space and constructing the dataset, the Extra Tree regression model and ANN (Artificial Neural Network) classification model are selected for prediction. The prediction accuracy for buckling load and mode shape reaches 98.34% and 93.75%, respectively. This significantly improves the prediction accuracy and efficiency, thereby overcoming the limitations of traditional methods.