A Transformer-based model to forecast traffic flow
Traffic flow forecasting plays a relative important role in operation of smart city. However, due to its high volatility and time-varying and regional vastness features, it could be difficult to precisely be predict. This research proposal a model to deal with traffic network, aiming at resolve the prediction problems in various kinds of traffic flow.
Predicton Performance Result shows here：
