Post-layout wiring violation prediction method based on deep learning and readable storage medium

The invention discloses a post-layout wiring violation prediction method based on deep learning and a readable storage medium, and the method comprises the steps: segmenting first layout information according to preset windows, so as to obtain first feature information corresponding to each preset window; obtaining a first feature image corresponding to each piece of first feature information according to the first feature information; obtaining a first five-dimensional tensor image according to all the first feature images corresponding to the same preset window; obtaining a corresponding first design rule violation image according to a preset window; obtaining a trained first network model; obtaining a first training model based on the trained first network model; obtaining a second training model; obtaining a final training model; and inputting the to-be-predicted data into the final training model to obtain a prediction result. According to the method, the layout can be adjusted intime under the condition