15results about How to "Save development time" patented technology

The invention discloses a data analysis and display system based on a large model, and the system comprises a data collection module which is used for carrying out the matching with a business database according to a current business stage, obtaining a matching result, starting a data burying point according to the matching result, and carrying out the data collection, and obtaining initial data; the data processing module is used for deploying a plurality of edge computing nodes on an edge server and performing multi-thread preprocessing on the initial data by utilizing the edge computing nodes; the data analysis module is used for transmitting the preprocessed initial data into an analysis model of the current service stage to obtain an analysis result; and the result display module is used for carrying out classified statistics on the preprocessed initial data and the analysis result to obtain a plurality of data reports and displaying the data reports in a monitoring center. The invention relates to the technical field of e-commerce data analysis, and solves the technical problems of low data processing efficiency and insufficient accuracy and breadth of data analysis in an existing e-commerce data analysis system.

The invention discloses an automatic way-finding method in a scene based on a ThreeJS framework, and relates to the technical field of webpage 3D application programs, and the method comprises the following steps: obtaining the size of a floor object in the scene on a three-dimensional coordinate system XOZ plane in advance, setting the unit length of a block grid unit to form a grid set, and setting the unit length of the block grid unit to form a three-dimensional coordinate system XOZ; the starting coordinate of the upper left corner of the grid set in the three-dimensional coordinate system, the coordinate of the center point of the first block grid unit and the column number and the row number of the grid set are calculated; according to the method, the optimal moving path between the two points is obtained in the scene based on the ThreeJS framework, the automatic path finding strategy in the 3D scene built by the ThreeJS is completed, the scene adaptability is high, the feasible path in the complex terrain can be accurately calculated according to the geometrical shape and object distribution of the scene, the role can freely move in the diversified terrain, and the effect of automatic path finding is achieved. According to the method, the path-finding accuracy and effectiveness are ensured, the method can be quickly integrated into a ThreeJS project, the automatic path-finding function is realized, and the development time and energy are saved.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing inference operations of machine learning models. One of the methods includes receiving, by a hardware device, data representing a machine learning model, the machine learning model including a plurality of model parameters for inference operations. A hardware device includes a set of computing units arranged in one or more processing elements. Instructions are obtained for performing an obfuscation operation configured to obfuscate one or more measurable characteristics of a machine learning model when the machine learning model is executed by one or more processing elements. A first portion of a set of computing elements is caused to perform an inference operation of a machine learning model, and a second portion of the set of computing elements is caused to perform an obfuscation operation concurrently with the first portion of the set of computing elements performing the inference operation.

The application provides a seamless integration method for components in Vue framework version evolution, belongs to the technical field of micro-frontend architecture and front-end component integration, and realizes decoupling and seamless integration of different versions of Vue components by loading old version Vue components into new version Vue applications through a micro-frontend container. The application can seamlessly use Vue 2 components in a Vue 3 project, reduces component library migration cost, avoids version conflicts, realizes quick integration of cross-version components and controllable management of the life cycle. The method has good compatibility, expansibility and high efficiency, and is suitable for version evolution and component reuse scenarios of large-scale front-end systems.

The embodiment of the invention provides a method and device for burying points in an application program, electronic equipment and a storage medium, and the method comprises the steps: carrying out the screenshot of a target page of the application program, and obtaining a screenshot page; the screenshot page and a view tree of the target page are reported to a service demand end, so that the service demand end positions the target component in the view tree of the target page by referring to the screenshot page, and a burying point position is obtained; acquiring burying point configuration data issued by the service demand end, wherein the burying point configuration data comprises a burying point position and a burying point event configured for the burying point position; and according to the point burying position and the point burying event, performing point burying on the target page to obtain a point burying page. In the embodiment of the invention, the business demand personnel of the business demand end can complete point burying through the visual screenshot and the view tree, the point burying is carried out after the application is online without the participation of developers, the point burying configuration is issued by the business demand end, the point burying adjustment can be carried out according to the business demand, the accuracy and comprehensiveness of the point burying are ensured, and the user experience is improved. And point burying is more flexible.

The invention belongs to the technical field of artificial intelligence, and relates to a page adaptation method and device based on an artificial intelligence model, computer equipment and a storage medium, and the method comprises the steps: obtaining different types of pages and equipment information; according to the page and the equipment information, training by using an artificial intelligence model; through machine analysis, disassembly and adaptation, semantic information and hierarchical relationships of page elements of the artificial intelligence model and adaptation rules on different devices are matched; recording page elements, components and adaptation rule information; calculating the size, the position and the display attribute of the page element according to the equipment information; and quickly generating an adapted page according to the calculated size, position and display attribute of the page element and the adaptation rule. The code repeated writing rate can be reduced, the development efficiency is improved, the development time and cost are saved, and the project online speed is increased; the compatibility of the page is enhanced, and the problems of page deformation, abnormal display and the like caused by compatibility difference are reduced.

The invention provides a cab noise prediction method and system, and relates to the technical field of vehicle strength analysis, and the method comprises the steps: obtaining a vibration sound transfer function from each cab suspension point to a driver position through different methods when there is a cab real object or a cab model, obtaining suspension point loads under different working conditions, and obtaining a cab noise prediction result; sound transfer functions from cab sound sources to the position of a driver are obtained through different methods, sound power level testing is carried out on each cab sound source, and sound source loads of the cab sound sources are obtained; combining the vibration-sound transfer function and the suspension point load to generate structural sound; generating air sound in combination with the sound transfer function and a cab sound source load; the cab noise is synthesized according to the structural sound and the air sound. According to the invention, the cab noise can be obtained more quickly and accurately.

The invention discloses a dynamic configuration method for a WeChat applet refrigerator control interface. The method comprises the steps that a refrigerator WeChat applet is started to select a refrigerator for control; sending a request to the cloud and carrying the refrigerator material number parameter; the cloud server returns a corresponding configuration file; analyzing the configuration file by the applet; dynamically generating a control interface; a user operates the functional assembly to control the refrigerator. The dynamic configuration method for the WeChat applet refrigerator control interface has the remarkable advantage that dynamic configuration of the refrigerator control interface is achieved. Only one set of applet codes needs to be developed, and control over multiple types of refrigerators is achieved through different configuration files.

The invention relates to a silicone oil clutch solenoid service life acceleration test method, which is characterized in that a set of acceleration test method for verifying the reliability or service life of a solenoid can be obtained through an algorithm according to a Cofen-Mansen model and a service life index of a customer; the reliability or life index of the solenoid is evaluated in an accelerated test mode, and the evaluated index can be used for guiding forward development of the solenoid; the reliability of the solenoid can be evaluated according to the reliability requirement of the OEM of the whole vehicle and the proportions of the road spectrum of the whole vehicle and the silicone oil clutch fan in different states, a simulation result equivalent to a simulation result under a long-term fatigue condition can be obtained in a short time, a large amount of time cost is saved, and verification cost is saved.

The invention provides a hot rolling method for obtaining non-bending-torsion deformation of complex-section profile steel, and relates to the technical field of optimization of profile steel rolling process parameters. The hot rolling method for obtaining the complex-section profile steel without bending-twisting deformation comprises the steps that a profile steel rolling whole-process finite element model is established; selecting a plurality of test factors, pre-selecting a plurality of level values for each factor, and designing a virtual orthogonal test; finite element calculation of each virtual orthogonal test is completed, the unit torsion angle and the bending degree per meter of each virtual test finished product are obtained, the two indexes are comprehensively scored, and the scoring result serves as a total index; normalizing each factor in a range of 0-1, and establishing a mathematical regression model between the orthogonal test total index y and each factor x; and with the maximum comprehensive score y of the torsion angle per unit length and the curvature per meter as a target and the upper and lower limits of the preselected level value of each factor x as constraint conditions, obtaining an optimal rolling process parameter combination for realizing the straightness of a finished product. The method is simple, convenient, safe, reliable, low in cost and high in production efficiency.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for obfuscating operations of neural networks. One of the methods includes receiving data representing a neural network. The neural network includes parameters specifying a sequence of network layers and a plurality of nodes in each layer of the sequence of network layers. The neural network is compiled to generate instructions that, when executed, cause one or more computing units of the hardware device to perform obfuscation operations associated with inference operations of the neural network. The obfuscation operation, when performed, obfuscates one or more measurable characteristics of the neural network. Compiling a target layer including determining a sequence of network layers; determining an obfuscated network structure for association with the target layer; and compiling the neural network and the associated obfuscation network structure together to generate an instruction for performing an obfuscation operation specified by the obfuscation network structure.

The invention discloses a battery module heat insulation pad type selection method and a test system, and belongs to the technical field of batteries, the battery module heat insulation pad type selection method comprises the steps that a to-be-tested battery module is acquired, the to-be-tested battery module comprises a plurality of battery cells and a plurality of heat insulation pads, the plurality of battery cells are arranged at intervals along a first direction, and the plurality of heat insulation pads are arranged at intervals along a second direction; a heat insulation pad is arranged between any two adjacent battery cells in the plurality of battery cells, the plurality of battery cells comprise a target battery cell, and the heat insulation capacities of the plurality of heat insulation pads are gradually increased along the direction far away from the target battery cell in the first direction; triggering thermal runaway of the target cell; and determining a thermal runaway propagation stop position in the direction away from the target cell. Thus, the heat insulation performance of a plurality of heat insulation pads can be tested at the same time through one test, the sample cost, the test cost and the development time are saved, and the test efficiency is greatly improved.

A kind of rail transit general vehicle controller and its control method, rail transit general vehicle controller includes automatic train protection unit, automatic train driving unit, driver train interface unit, vehicle input and output unit, speed measurement distance unit, transponder transmission unit, track circuit information receiving unit, data recording unit and wireless communication unit by high-speed real-time communication bus with each other.The general vehicle controller is run in single mode control mode, the protection unit in the automatic train protection unit that matches the current operating line mode is as the current main control protection unit, and the current main control protection unit exclusively accesses other equipment unit;General vehicle controller is run in multimode control mode, the protection unit in the automatic train protection unit that matches the current operating line mode is as the current main control protection unit in turn, and the current main control protection unit shares access other equipment unit.The vehicle-mounted product architecture of the application is unified, product reuse degree is high, easy to extend.

The present application provides a method and device for parsing communication instructions of a sorter for final test equipment. The method is executed by the final test equipment connected to the sorter, and the communication instruction records one of the following items: the occupation status of at least one station, the bin value of the device corresponding to at least one station; the method includes: extracting the same content of the first communication instruction obtained and the second communication instruction; extracting the different content of the first communication instruction obtained and the second communication instruction obtained; according to the same content and the different content, parsing the encoding method of the characters used to indicate the occupation status of the station or the encoding method of the characters used to indicate the bin value of the device corresponding to the station in the communication instruction. The method can parse various types of communication instructions of sorters for final test equipment, and is used to support sorters of various types of communication instructions, with good reusability and high operating efficiency, saving the workload and development time of developing drivers corresponding to new types of instructions.

This utility model relates to a pressure relief valve structure for an engine oil pump, including an oil pump housing and a pressure relief valve assembly. The connection between the oil pump housing and the pressure relief valve assembly is provided with an internal thread. The oil pump housing is threadedly connected to the external thread of the valve stem on the pressure relief valve assembly via the internal thread. The oil pump housing and the pressure relief valve assembly are sealed by a top sealing ring and a bottom sealing ring fitted on the pressure relief valve assembly. This utility model solves the problem of oil pump pressure relief valve jamming; avoids risks such as lubrication system seal leakage and engine cylinder scoring; and saves oil pump development time and costs.