The acquisition of the initial dynamic vivacity ratio of the propellant charge requires dynamic compression and fracture of the propellant charge and initial dynamic vivacity ratio testing, which is time-consuming and difficult to predict in advance. Whether it is possible to directly obtain the dynamic compression and fracture test conditions of the propellant charge under the corresponding initial dynamic vivacity ratio is related to the improvement of test efficiency. To this end, a multi output support vector regression model suitable for small sample data regression was adopted. Three multi output support vector regression models were established successively to predict the dynamic compression and fracture test conditions of propellant charge. The prediction results were verified by experiments, demonstrating the superiority of the established multi output support vector regression model in predicting unknown datasets under small sample data. This provides an efficient and accurate method for predicting the dynamic compression and fracture test conditions of propellant charge under corresponding initial dynamic vivacity ratios.