This paper aimed to investigate the antibacterial activity of flowers and leaves from Paeonia rockii, screen antibacterial compounds and predict targets of antibacterial to explore its multi-component, multi-target antibacterial mechanism. In this study, minimal inhibitory concentration(MIC) of seven strains of Staphylococcus aureus, S. epidermidis， Bacillus subtilis, B. cereus, Escherichia coli, Klebsiella pneumoniae and Pseudomonas aeruginosa were determined by microdilution method. Uniprot databases was used to find the antibacterial targets, and RCSB was used to identify targets associated with antimicrobial activity as docked receptor proteins. The candidate active ingredients from flowers and leaves of P. rockii were identified by database such as PubChem. The ligands were constructed by ChemDraw, Avogadro and Discovery Studio Visualizer. QuickVina 2.0 software was used to molecular docking. Besides, the Cytoscape 3.5.1 software was used to construct activity compounds of flowers and leaves from P. rockii ingredients-targets network, and Uniprot software was used to analyze gene ontology and KEGG pathway. In vitro antibacterial experiments found antibacterial effect of the flowers and leaves from P. rockii, especially methanol extraction of flowers has the strongest antibacterial effect. The network pharmacology indicated that total 29 activity ingredients and their 18 targets were screened in flowers and leaves from P. rockii. Comparison of the active ingredients and the number of antimicrobial target networks, it is predicted that the antibacterial components are mainly flavonoids and phenolic acids and main mechanism of antibacterial is to inhibit the synthesis of bacterial proteins. In this study, potential antibacterial activity of flowers and leaves from P. rockii has be found by antibacterial experiments in vitro and network pharmacology screening. And this study provides new clues for further basic study on the antibacterial agents of flowers and leaves from P. rockii.