The Traffic Flow Prediction Method Using the Incremental Learning-Based CNN-LTSM Model: The Solution of Mobile Application

Algorithm 1

IL-TFNet model training algorithm.

	Input: , represents the number of samples, , .
	Output: IL-TFNet model with certain predictive performance after training.
	Phase 1: Initialization phase:
	(1) Training and testing data set initialization: Divide traffic flow data set D into multiple subtraining data sets (D1, D2, ..., Dn − 1) and one testing data set (Dn).
	(2) Initialization iterations , Learning rate , Attenuation rate , .
	(3) Initialize first-order momentum and second-order momentum , and the parameter to ensure numerical stability .
	(4) Initialize the CNN parameters of IL-TFNet model: , , .
	Phase 2: Incremental training phase:
	(1) while current subtraining data set is not empty do
	(2) while the termination condition is not reached (the model prediction performance is in a bottleneck or reaches a certain iteration round) do
	(3)  Sample the subtraining data set to obtain current batch training data set
	(4)  Calculate gradient based on formula (7):
	(5)  Update iterations
	(6)  Update first-order momentum and second-order momentum: ,
	(7)  Calculate bias correction ,
	(8)  Update model parameters
	end while
	(9) Use the next subtraining data set to train the model in turn
	end while
	(10) return (Model parameters after training)