Convolutional neural network model compression method combining pruning and knowledge distillation

The invention relates to a convolutional neural network model compression method combining pruning and knowledge distillation. The method comprises the following steps: acquiring an image training set A; obtaining a target network model, and introducing a scaling factor gamma to each channel contained in the target network model; training the target network model, and taking the trained model as a teacher network; pruning the channel number of the teacher network according to the absolute value of the scaling factor gamma, and regarding the pruned model as a student network; acquiring a small number of images in the image training set A, inputting the images into teacher and student networks, and respectively calculating distribution differences between feature maps output by each convolutional layer channel of the teacher and student networks; using the distribution difference as a loss function to train the student network, so that the model precision of the student network is quickly recovered to the level