Self-adaption method, system and device during edge cloud collaborative model testing and medium

The invention discloses a self-adaption method and device during edge cloud collaborative model testing and a storage medium, and belongs to the technical field of edge computing. The method comprises the following steps: each edge device executes a model reasoning operation, obtains and uploads metering information and logics and outputs the metering information and logics to a cloud end; the cloud carries out joint estimation on the information uploaded by the edge device; the cloud generates a pseudo sample according to the joint statistical information data of each edge device; the cloud terminal optimizes the model parameters according to the pseudo samples and issues the optimized model parameters to the edge device; and the edge device updates a local model according to the model parameters issued by the cloud, and carries out subsequent sample reasoning operation. According to the method, model self-adaption is carried out between the cloud and the edge device in a cooperative manner, so that extra ca