Traffic flow prediction method and device based on space-time sequence deep learning

The invention relates to a traffic flow prediction method and device based on time-space sequence deep learning. The method comprises the following steps: obtaining a road shape and traffic flow historical data of a road; the method comprises the following steps: preprocessing traffic flow historical data according to a related time correlation sequence, further arranging the traffic flow historical data into a tensor form according to a batch form, and constructing a multivariable time-space sequence data set of the traffic flow historical data; dividing the multivariable space-time sequence data set into a training data set, a verification data set and a test data set; training the traffic flow prediction model by using the training data set; and collecting traffic flow data at the current moment, inputting the collected traffic flow data at the current moment into the trained traffic flow prediction model, and predicting a time sequence value at a future moment. According to the method, the effective chang