12results about How to "Reduce the time spent" patented technology

The embodiment of the invention discloses a preprocessing method and device applied to a graph neural network, equipment, a medium and a product, and the method comprises the steps: determining the node information of a sampling node corresponding to each processing unit in a graphics processor according to at least one sub-graph associated with the graph neural network; for each processing unit, according to the network structure of the graph neural network and the corresponding sub-graph structure, determining a target remote node and a target local node executed by each thread by taking the quantity of remote operations and local operations processed by each thread associated with the processing unit to meet a preset condition as an optimization target; and configuring a feature acquisition operator and a sampling operator for each thread, acquiring features corresponding to the target remote node and the target local node based on the feature acquisition operator, and sampling the target remote node and the target local node based on the sampling operator. According to the technical scheme, the preprocessing efficiency of the graph neural network is improved while the performance of the graph neural network model is not affected.

The invention discloses a device for automatically recognizing pictures, characters and sound in real time, which comprises a data acquisition module used for acquiring picture, character or sound information of a specified target; the extraction module is used for extracting knowledge points in the picture, character or sound information; the knowledge base module is used for storing knowledge data information; the comparison module is used for comparing the knowledge points extracted by the extraction module with the knowledge data information in the knowledge base module to obtain comparison information; and the correcting module is used for correcting the knowledge points of the specified target according to the comparison information. Compared with the prior art, characters, pictures and sounds can be automatically recognized, the purpose of automatically correcting homework is achieved, and therefore the correcting time of teachers can be shortened.

The utility model provides a rail electric movable type dustproof shed for GIS installation, which comprises a dustproof shed frame, large electric crane running end beams are arranged on the left side and the right side of the dustproof shed frame, the bottoms of the large electric crane running end beams are connected with light rail tracks in a sliding mode, an electric folding sunshade is arranged on the top of the dustproof shed frame, and the electric folding sunshade is connected with the large electric crane running end beams in a sliding mode. Dustproof awning cloth is arranged on the left side and the right side of the dustproof awning frame, the dustproof awning cloth is sectional type awning cloth and is provided with a plurality of cuffs and door openings, and a plurality of hanging rings are arranged on the top of the electric folding sunshade. The dustproof shed is simple in structure and reasonable in design, is of a spliced structure, can be quickly assembled on site, is very simple to assemble and disassemble, can be conveniently connected with an air shower, a changing room, an air conditioner and GIS (Gas Insulated Switchgear) related equipment, and is good in sealing performance.

The invention discloses a standby vehicle inspection method and device for a commercial vehicle and an electronic product. The method comprises the steps that vehicle preparation plan information of a target vehicle preparation lane is acquired, and the vehicle preparation plan information comprises at least one of the following information: a first vehicle preparation number of planned vehicle preparation of the target lane, vehicle information of each first target commercial vehicle and target vehicle preparation duration; acquiring acquisition information of the target standby lane, wherein the acquisition information at least comprises one of vehicle information of each second target commercial vehicle passing through the target lane and standby vehicle time of each second target commercial vehicle passing through the target standby lane; determining standby vehicle actual information of the target standby vehicle lane according to the standby vehicle plan information and the collection information; verifying the standby vehicle actual information and the standby vehicle plan information, and generating a verification result; and determining whether the vehicle preparation plan is completed or not according to the verification result. According to the invention, the technical problems of low inspection efficiency, long time consumption and inaccurate inspection result caused by manual completion of standby vehicle inspection are solved.

The application relates to an AI-driven conversational ERP order generation method and system, and belongs to the field of resource planning systems. The method comprises the following steps: loading user permissions and collecting historical behavior sequences in response to a wake-up instruction; predicting an order intention and generating an active guidance prompt based on a time sequence coding model; extracting initial order elements by using a large language model, combining a static permission vector and a dynamic context vector for joint constraint optimization to generate accurate order elements; calling an ERP master data interface to obtain real business data after permission verification to generate a basic order data set; performing conflict resolution through a structured constraint rule library and a model output-rule bidirectional verification mechanism to generate a pre-order data object; and writing into a database to generate an official order after multi-modal fusion confirmation and risk control pre-checking. The application has the advantages that permission verification is pre-positioned, data is accurate and reliable, active guidance is provided, and multi-modal risk pre-checking is performed, so that order entry efficiency and compliance are significantly improved.

The invention provides a GNSS partial ambiguity fixed acceleration method based on isomorphic multiplexing, and the method comprises the steps: carrying out the first complete decorrelation calculation of a covariance matrix of a first ambiguity subset, and systematically recording the decorrelation transformation information of the first ambiguity subset; when a subsequent ambiguity subset is processed, pre-transforming the covariance matrix of the current subset into a state closer to a decorrelation characteristic by utilizing the structure isomorphism between the covariance matrix of the current subset and the covariance matrix of the previous subset and multiplexing the recorded transformation information, and taking the state as an efficient starting point of iterative optimization; therefore, completely independent and time-consuming correlation reduction operation is prevented from being carried out on each subset covariance matrix from zero. According to the method, the overall calculation burden of sequential drop correlation in the partial ambiguity fixing process is remarkably reduced, the calculation efficiency is greatly improved, and the method is particularly suitable for a low-cost GNSS receiver and other high-real-time application platforms which have strict real-time requirements or are limited in calculation power.

This utility model discloses a preserved flower bonding device, relating to the technical field of preserved flower processing equipment. The device includes a base, with a vertically arranged column rotatably mounted on the upper surface of the base. A vertically arranged fixing column is movably inserted into the top of the column. A drive motor is fixedly mounted inside the base below the column, with its output end coaxially fixed to the column. Three annularly distributed limiting strips are fixedly mounted on the outer wall of the fixing column, inserted along with the fixing column. A horizontally arranged transverse plate is fixedly mounted on the top of the fixing column, with a groove on its upper surface. Two symmetrically arranged positioning plates slide within the groove. In this utility model, the process of the drive motor rotating the column and the electric push rod pushing the positioning plates upwards requires only simple operations such as "feeding material and applying glue," eliminating the need for continuous force to control the connection and significantly reducing physical burden.

The invention provides a multi-operation-mode section power transmission capacity evaluation method, device and equipment and a medium, and relates to the technical field of power systems, and the method comprises the steps: building a multi-operation-mode model based on the operation parameters of power equipment; according to the multi-operation mode model, constructing a main problem optimization model which aims at minimizing the operation cost of the power system and maximizing the power transmission capacity of the section to be evaluated; according to the optimization target of the main problem optimization model, designing a double-layer robust sub-problem model for quantifying the reliability margin of the to-be-evaluated section; iteratively solving the main problem optimization model and the double-layer robust sub-problem model to obtain an available power transmission capacity evaluation model; a multi-target clustering algorithm is adopted to classify the multi-operation-mode sample set, and clustering agent points and probability weights of the clustering agent points are obtained; and integrating the evaluation model and the clustering result, and outputting an evaluation result of the to-be-evaluated section. According to the method, the hostile scene is captured through the double-layer robust sub-problem model, and the accuracy of the obtained evaluation result is improved.

The present invention proposes a parallel detection method, platform, system, device and medium for power distribution terminals, and relates to the technical field of power distribution terminal detection. In a parallel detection method for power distribution terminals proposed by the present invention, based on the separate communication addresses of each hardware device control module, decoupled communication interactions are performed respectively, so that the detection process control module can perform precise control on each hardware device control module; by monitoring whether the data transmitted by each tested power distribution terminal is lost, the tested power distribution terminal tested by the corresponding hardware device control module is bypassed when the data is lost, so as to reduce the interference to other hardware device control modules tested in parallel, and finally perform automated and parallel testing of multi-process decoupling control, so as to reduce the impact and time consumption caused by sudden abnormalities in the detection process.

The embodiments of the present application disclose a data identification method, device, electronic device and storage medium; including: obtaining multiple classifier sets, each classifier set includes at least two classifiers; for multiple historical time periods, obtaining the number of hits of the text data corresponding to each historical time period to each classifier set; if the hit difference between the number of hits of the text data corresponding to the first historical time period to the first classifier set and the number of hits of the text data corresponding to the second historical time period to the first classifier set exceeds a preset threshold, then the first classifier set is determined to be an abnormal classifier set; and at least one abnormal classifier set is displayed. The embodiments of the present application can screen and display abnormal classifier sets, and the staff can quickly summarize the corresponding abnormal data after viewing, which can reduce the consumption time and reduce the consumption of human resources.

This utility model discloses a portable, universal hydraulic cylinder disassembly device for field use, belonging to the field of hydraulic cylinder repair technology. It includes a main body with a lifting assembly at the top. The lifting assembly includes a lifting frame vertically slidably mounted within the main body, and a torque converter at the bottom. Clamping assemblies are connected to both the lifting frame and the torque converter. The two clamping assemblies are used to clamp cylinder head A and cylinder barrel B respectively. The torque converter drives the connected clamping assemblies to twist cylinder head A or cylinder barrel B. Compared with existing technologies, this portable, universal hydraulic cylinder disassembly device eliminates the need for manual disassembly of hydraulic cylinders using tools or large horizontal disassembly machines, reducing the time spent disassembling hydraulic cylinders and avoiding deformation and damage caused by uneven force on the hydraulic cylinder. It also eliminates the traditional step of disassembling by hammering.

The embodiments of the present application disclose a method, device and electronic device for obtaining feature information of a target object. The method includes: the client obtains the size information of at least one available template, the size information includes the width information of the available template and the height information of the available template; obtains the search area corresponding to the target object to be tracked in the current frame image, creates a corresponding storage area for the search area, and saves the image information associated with the search area to the storage area; determines the target template, and scales the image information in the storage area according to the size information of the target template to generate a template image; extracts feature information related to the target object from the template image, so as to track the target object in the next frame image according to the feature information. Such a scheme reduces the time complexity of the algorithm while minimizing the impact on the tracking effect of the algorithm.