Experimental data modeling evaluation method and system

The invention relates to the technical field of automatic experiments, discloses an experimental data modeling evaluation method, and particularly relates to an experimental data modeling evaluation method and system. In the technical scheme provided by the embodiment of the invention, a basic evaluation network is constructed, and the basic evaluation network is optimized through a recursive regularization algorithm to obtain an optimized target evaluation network. The basic evaluation network is composed of a multi-layer basic evaluation sub-network, a loss function layer and a full connection layer, the multi-layer basic evaluation sub-network is optimized, and the optimized multi-layer basic evaluation sub-network is connected with the loss function layer and the full connection layer to form a final target evaluation network. According to the method, the evaluation result of the experimental data is obtained through the constructed multi-layer basic evaluation sub-network, and compared with the prior art