Abstract: Embodiments of this disclosure provide a method and apparatus for compiling a neural network model, and a method and an apparatus for training an optimization model. The method includes: obtaining a to-be-compiled neural network model; determining an intermediate instruction sequence corresponding to the to-be-compiled neural network model based on the to-be-compiled neural network model; processing the intermediate instruction sequence by using a pre-trained instruction sequence optimization model, to obtain a target optimization parameter corresponding to the intermediate instruction sequence; determining an optimization instruction sequence corresponding to the intermediate instruction sequence based on the target optimization parameter; and converting the optimization instruction sequence into an executable instruction sequence, to obtain a target instruction sequence that is executable by a neural network processor corresponding to the to-be-compiled neural network model.

Abstract: A mass flow controller includes control module, fluid passage, flow regulating valve, flow sensor disposed between inlet and flow regulating valve, and pressure sensor disposed at outlet. The control module is configured to enter pressure feedback regulation mode when variation amount of target flow value exceeds preset threshold, and enter flow feedback regulation mode after fluid flow value detected by flow sensor meets first stabilization condition. The control module is configured to, in pressure feedback regulation mode, perform calculation according to fluid pressure value and target flow value to obtain opening regulation amount, and regulate opening degree of flow regulating valve according to opening regulation amount. The control module is also configured to, in flow feedback regulation mode, perform calculation according to fluid flow value and target flow value to obtain opening regulation amount, and regulate opening degree of flow regulating valve according to opening regulation amount.

Abstract: A mass flow controller includes a control module (300), a fluid passage, a flow regulating valve (200), a flow sensor (400) disposed between an inlet and the flow regulating valve (200), and a pressure sensor (500) disposed at an outlet. The control module (300) is configured to enter a pressure feedback regulation mode when a variation amount of a target flow value exceeds a preset threshold, and enter a flow feedback regulation mode after a fluid flow value detected by the flow sensor (400) meets a first stabilization condition. The control module (300) is configured to, in the pressure feedback regulation mode, perform a calculation according to a fluid pressure value and the target flow value to obtain an opening regulation amount of the flow regulating valve (200), and regulate an opening degree of the flow regulating valve (200) according to the opening regulation amount.

Abstract: This disclosure relates to a tire self-positioning system and a positioning method. The tire self-positioning method includes the steps of: S1, performing data acquisition; S2, performing data transmission; S3, performing data conversion; S4, repeatedly executing steps S1 to S3; S5, performing data statistics; S6, determining, according to a statistical result, position information of a tire corresponding to a first signal. According to the tire self-positioning system and the tire self-positioning method, the accuracy of self-positioning can be improved.

Abstract: Methods and systems for detecting fraudulent activity at a self-checkout terminals of a retail store include a scanner for scanning an identifier of a candidate product located in the product-scanning area of the self-checkout terminal, and one or more sensors that detect at least one physical characteristic of the candidate product located in the product-scanning area of the self-checkout terminal. A computing device then correlates the obtained electronic data corresponding to actual identifying characteristic information associated with the candidate product to the reference physical characteristic information associated with the reference product in order to generate a similarity score between the actual and reference physical characteristic information.