The invention belongs to the technical field of
traffic prediction, and particularly discloses a continuously memorized adaptive heterogeneous space-time diagram
convolution traffic prediction method and
system.The method comprises the steps that
traffic flow data and historical memories of flow data are input into a memory input layer, and the memory input layer outputs a
time sequence; the
time sequence is used as the input of a first sub-layer of a heterogeneous space-time diagram
convolution layer, the heterogeneous space-time diagram
convolution layer is provided with a plurality of sub-
layers, the output of the previous sub-layer is the input of the next sub-layer, different space-time heterogeneous diagrams are constructed, and the space-time heterogeneous diagrams are used for completing the diagram convolution operation. And each layer of the heterogeneous space-time diagram convolution layer outputs a
time sequence to the space-
time information fusion layer to obtain
traffic flow prediction data and a new historical memory. By adopting the technical scheme, the heterogeneity of the
traffic flow data is captured, the long-term dependence of the traffic flow is obtained through the historical information, and the prediction effect is improved.