Method and apparatus for warning

An apparatus for warning, comprising:

• at least one processor;
• and
• a memory storing instructions, which when executed by the at least one processor, cause the at least one processor to perform operations comprising:
• importing a real-time collected traffic data sequence into a pre-trained traffic prediction model, to obtain predicted traffic data, the pre-trained traffic prediction model being used to calculate and obtain the predicted traffic data according to the real-time collected traffic data sequence;
• and
• sending, in response to a difference between the predicted traffic data and actual traffic data corresponding to time being greater than a set threshold, a warning signal;
• wherein the pre-trained traffic prediction model comprises a convolutional neural network, a residual network and a fully connected layer;
• and
• wherein the residual network comprises a deep characteristic extraction network and a surface characteristic extraction network, wherein the deep characteristic extraction network of the residual network comprises at least two convolutional layers, and an output of a previous convolutional layer in the at least two convolutional layers is used as an input of a later convolutional layer adjacent to the previous convolutional layer.

A non-transitory computer readable medium, storing a computer program, wherein the program, when executed by a processor, causes the processor to perform operations comprising:

• importing a real-time collected traffic data sequence into a pre-trained traffic prediction model, to obtain predicted traffic data, the pre-trained traffic prediction model being used to calculate and obtain the predicted traffic data according to the real-time collected traffic data sequence;
• and
• sending, in response to a difference between the predicted traffic data and actual traffic data corresponding to time being greater than a set threshold, a warning signal;
• wherein the pre-trained traffic prediction model comprises a convolutional neural network, a residual network and a fully connected layer;
• and
• wherein the residual network comprises a deep characteristic extraction network and a surface characteristic extraction network, wherein the deep characteristic extraction network of the residual network comprises at least two convolutional layers, and an output of a previous convolutional layer in the at least two convolutional layers is used as an input of a later convolutional layer adjacent to the previous convolutional layer.