3results about How to "Reduce memory pressure" patented technology

This application relates to a data storage method, apparatus, vehicle, electronic device, and storage medium. The method includes: determining the computation order of N operators included in a neural network; determining M operators from the N operators based on the computation order; calculating the i-th operator among the N operators according to the computation order to obtain the output of the i-th operator; if the i-th operator is one of the M operators, then saving the output of the i-th operator to a first type of storage space, wherein the output of the i-th operator saved in the first type of storage space cannot be deleted after calculating the (i+1)-th operator but before calculating the k-th operator; if the i-th operator is not one of the M operators, then saving the output of the i-th operator to a second type of storage space, wherein the output of the i-th operator saved in the second type of storage space can be deleted after calculating the (i+1)-th operator. This method can solve the problem of excessive memory consumption during the computation of a neural network model.

A neural network inference acceleration method and device are provided. The method includes: acquiring, by a memory having a computing unit, a video and a prompt word; acquiring, by the memory having the computing unit, at least one image related to the prompt word from the video; dividing, by the memory having the computing unit, each of the at least one image into a plurality of blocks; storing, by the memory having the computing unit, the plurality of blocks; storing, by the memory having the computing unit, position information of the plurality of blocks in each of the at least one image; acquiring, by a processor, each of the at least one image as the plurality of blocks and the position information; dividing, by the processor, each of the at least one image into a plurality of sub-blocks; acquiring, by the processor, from the memory having the computing unit, an embedded vector of a pre-stored sub-block corresponding to each of the plurality of sub-blocks, respectively; and performing an inference operation using the embedded vector and a first neural network.

This application provides an application installation method, apparatus, device, and medium based on compilation optimization, relating to the field of computer technology. The method includes: in response to an installation operation on a target application, obtaining an application package corresponding to the target application; extracting baseline configuration files for various scenarios included in the application package; installing the target application based on the application package, and compiling hotspot code corresponding to a first configuration file that meets the compilation requirements during installation from the baseline configuration files for each scenario; and, when it is determined that the electronic device is in an idle state, compiling hotspot code corresponding to a second configuration file (excluding the first configuration file) from the baseline configuration files for each scenario, thereby completing the installation of the target application. This method aims to save resources, improve performance, and enhance the user experience.