With the construction of large-capacity long-distance high-voltage direct current transmission projects and the large-scale integration of renewable energy, the frequency security of the power system is facing severe challenges. For fast and accurate online assessment of frequency security, a data-driven frequency security assessment model based on Generative Adversarial Network(GAN) and Metric Learning(ML) is proposed. Firstly, the key frequency security indicators are selected as the outputs of the model, and the input feature set is constructed; Then, the Wasserstein Generative Adversarial Network(WGAN) technique based on Wasserstein distance metric is applied to learn the distribution information of historical operation scenarios of power systems, to generate operation scenarios covering typical operation modes to build the training sample set; Finally, considering the inapplicability of a single machine learning model to frequency security assessment under complicated operation modes of power systems, a combined assessment model for frequency security assessment composed of multiple sub-models is constructed based on Metric Learning for Kernel Regression (MLKR) method. Finally, a simplified Shandong power system example is used to verify the effectiveness of the proposed method.