Multi-modal classification model generation method and device

The invention provides a multi-modal classification model generation method and device, and relates to the technical field of artificial intelligence. One specific embodiment of the method comprises the following steps: acquiring a preset sample set, wherein a sample comprises at least two sub-samples with different modals; obtaining a pre-established multi-modal fusion network, wherein the multi-modal fusion network comprises a threshold module, a modal fusion module and at least two classification models for classifying different modal data; selecting samples from the sample set; respectively inputting sub-samples of different modals of the sample into classification models corresponding to the modals to obtain feature vectors output by the classification models, extracting threshold vectors of all the feature vectors through a threshold module, and inputting all the feature vectors and the threshold vectors into a modal fusion module to obtain a modal fusion model; and if the multi-modal fusion network mee