Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

68results about How to "Improve iteration efficiency" patented technology

Algorithm model verification and precision evaluation method and device, equipment and storage medium

The invention belongs to the field of artificial intelligence, finance and medical treatment, and relates to an algorithm model verification and precision evaluation method, which comprises the steps of calling a pre-constructed original algorithm model to process an original sample file set to obtain a to-be-evaluated model result, and associating an original sample file with the to-be-evaluated model result to obtain a to-be-evaluated model result; performing annotation task allocation according to the generated annotation task allocation scheme, receiving a truth value result submitted by the annotation personnel, comparing a to-be-evaluated model result with the corresponding truth value result to obtain an initial model precision evaluation result, optimizing the model according to the initial model precision evaluation result, and evaluating the optimization algorithm model to obtain an evaluation result. And generating a model performance comparison report. The invention further provides an algorithm model verification and precision evaluation device, equipment and a storage medium. In addition, the invention also relates to a block chain technology, and the original sample file can be stored in a block chain. According to the method, the model iteration efficiency can be greatly improved, and the precision and performance of the model are improved.
Owner:PING AN TECH (SHENZHEN) CO LTD

Speech synthesis model training method and device, speech synthesis method and device and storage medium

The invention provides a speech synthesis model training method, a speech synthesis method, a speech synthesis device and a storage medium, and the speech synthesis model training method comprises the steps: obtaining the sample text data of a plurality of speakers and the acoustic features of the sample audio data of the plurality of speakers according to the sample audio data of the plurality of speakers, feature extraction is carried out on the sample text data of each speaker to obtain sample text features of each speaker, a speech synthesis sample of each speaker is constructed according to the sample text features of each speaker and acoustic features of the sample audio data of each speaker, and speech synthesis is carried out according to the speech synthesis samples of multiple speakers. And respectively carrying out model training to obtain speech synthesis models of the plurality of speakers. Automatic training of the speech synthesis model is carried out in a one-button mode, and the iteration efficiency of the speech synthesis technology is improved.
Owner:NETEASE (HANGZHOU) NETWORK CO LTD

Power equipment asset management method and system

The invention relates to the technical field of data processing, in particular to a power equipment asset management method and system. The method comprises the following steps: determining the selection probability of each dimension under child nodes according to the distribution of power data in the child nodes of an isolated tree of power equipment, determining the segmentation necessity of the child nodes according to the time sequence regularity and the selection probability of the power data in the child nodes and the number of the power data in the child nodes, and when the segmentation necessity is greater than a segmentation threshold value, determining the segmentation necessity of the child nodes; according to the selection probability, randomly selecting one dimension in an unequal probability manner to divide the sub-nodes, continuously updating the isolated forest according to the newly generated power data, identifying the current state of the power equipment according to the abnormal score of the current power data, and synchronizing the current state of the power equipment to the digital twin model of the power equipment in real time. And power equipment asset monitoring and management are realized. According to the invention, the state identification of the power equipment is accurate and efficient, and the real-time performance of power equipment asset monitoring and management is improved.
Owner:STATE GRID HUBEI ELECTRIC POWER INFORMATION & TELECOMMUNICATION COMPANY +2

Intelligent mold locking device and intelligent mold locking method

The invention discloses an intelligent mode locking device and an intelligent mode locking method. The intelligent mode locking device comprises a programmable control pumping source; the wavelength division multiplexing coupler is connected with the programmable control pumping source; the thulium-doped gain optical fiber is connected with the wavelength division multiplexing coupler; the first coupling output device is connected with the thulium-doped gain fiber, the first coupling output device is connected with the second coupling output device, and the second coupling output device is connected to the photoelectric detector and the spectrometer; the unidirectional polarization dependent isolator is connected with the first coupling output device; one end of the electric polarization controller is connected with the one-way polarization dependent isolator, and the other end of the electric polarization controller is connected with the wavelength division multiplexing coupler; the photoelectric detector is connected with the second coupling output device; the oscilloscope is connected with the photoelectric detector; the spectrograph is connected with the second coupling output device; and the mode locking control module is connected with the programmable control pumping source, the electric polarization controller, the oscilloscope and the spectrograph.
Owner:SHENZHEN UNIV

Software integration method, device and equipment, storage medium and vehicle

The invention relates to a software integration method and device, equipment, a storage medium and a vehicle, and the method comprises the steps: obtaining first component information of a current version of target software and second component information of a previous version, the target software comprises a plurality of development modules, and different component information comprises different development module labels and configuration information; the configuration information is compared, a target configuration environment is determined according to a comparison result, the development module labels are compared, and the multiple development module labels belonging to the same development module are compared one on the year by year; determining a third development module label with difference and a fourth development module label without difference from the first development module labels according to a comparison result; the first development module corresponding to the third development module label is recompiled, the second development module corresponding to the fourth development module label is controlled to reuse the compiling product of the last version of the target software, and the efficiency of the iteration process is improved, and the time cost is saved.
Owner:CHONGQING CHANGAN AUTOMOBILE CO LTD

Reaction spectrum reverse iteration algorithm with additional damping ratio

The invention discloses a reaction spectrum reverse iteration algorithm of an additional damping ratio, relates to the technical field of constructional engineering, and aims to reversely calculate reasonable parameters of a damper and a connecting component from the final demand (namely, the additional damping ratio demand and the rigidity demand of a damping target) of a main body structure, and merge and iterate according to the model of a common damper product to obtain a response spectrum reverse iteration algorithm of the additional damping ratio. And compared with the traditional forward iteration, the efficiency of the reverse iteration algorithm is higher.
Owner:NO 1 ENG CO LTD OF FHEC OF CCCC +1

Fuel power calibration method for loop irradiation test

ActiveCN119442954ARealize OKEliminate the effects of uncertaintyNuclear energy generationNuclear monitoringNuclear powerBurnup
The invention relates to the technical field of research of reactor fuel irradiation, in particular to a fuel power calibration method for a loop irradiation test, which comprises the following steps: carrying out an out-of-reactor hydraulic test, and establishing a high-precision CFD (computational fluid dynamics) model of a fuel loop irradiation device based on a hydraulic test result; calculating the gamma heat release rate ratio of each material and the total heat release in the test section, and preliminarily determining the nuclear power value of the fuel assembly; based on the heat conduction differential equation and the convective heat transfer equation, the gamma heat release rate of the material and the nuclear power value of the fuel assembly are corrected; repeating the calculation process of S3 until the nuclear power value of the fuel assembly converges, and obtaining the final theoretical calculation value of the gamma heat release rate of each material and the nuclear power of the fuel assembly; and further verifying the calculation result to finally obtain the nuclear power of the fuel core body, and obtaining the burnup of the fuel assembly in the irradiation test through time integration. The method aims to solve the technical problem that the nuclear power of the fuel assembly calculated based on thermotechnical measurement data in a loop irradiation test has deviation compared with an actual value.
Owner:NUCLEAR POWER INSTITUTE OF CHINA

Financial report generation method and system, electronic equipment and storage medium

The invention discloses a financial statement generation method and system, electronic equipment and a storage medium. The generation method comprises the steps of obtaining a general large model; generating a private model by combining an external pre-selection platform based on the general large model; the private model is used for generating a financial statement; obtaining a low-code platform, integrating the private model with the low-code platform, and constructing a target platform; and obtaining a user demand, and constructing a financial statement based on the target platform and the user demand. The process of AI model data processing and model training and deployment is simplified, developers can easily construct an interactive AI model training and deployment interface through a low-code platform, visual customization and batch generation of AI private large models are completed, the difficulty of AI model development and deployment is lowered, and the development efficiency is improved. And the technical threshold of customizing the model is reduced and the efficiency of model iteration is improved, so that the efficiency and the accuracy of financial statement generation are improved.
Owner:SHANGHAI LIANGTOU NETWORK TECH CO LTD

Dynamic route replacement method, system and equipment based on micro front end and medium

The invention provides a dynamic route replacement method, system and device based on a micro front end and a medium, which are applied to a micro front end container running in a browser, and the method comprises the following steps: receiving an access request sent by a user; the access request is a uniform resource path; acquiring a set routing mapping table in the micro front-end container; the set routing mapping table comprises first routing information and second routing information; determining a corresponding accessible route through a logic matching rule according to the access request and the set route mapping table; and loading and displaying the corresponding sub-application according to the accessible route. Compared with the traditional technical scheme, the method has the advantages that the route dynamic replacement logic is moved forward into the web container, so that the complexity of operation and maintenance deployment can be greatly reduced, and the problems of tedious maintenance and high maintenance cost are solved.
Owner:SHANGHAI HUIYUN XINZHI TECHNOLOGY CO LTD

Rectangular aperture curved surface shape error compensation method and system

The invention provides a rectangular aperture curved surface shape error compensation method and system, and belongs to the field of free-form surface optical element ultra-precision machining.The method comprises the steps that the surface of a workpiece is machined based on an optimal moving path, coordinates and actual height values of all points on a curved surface shape are collected, and machining errors are obtained; performing unit square normalization on the processing error curved surface by using a Zernike polynomial, and then fitting a prediction error curved surface; and the predicted error curved surface is negated to correct the optimal moving path in the machining process, and the surface of the workpiece is remachined till the machining error meets the machining precision. The method strictly controls the precision and range of error compensation, and ensures that the compensated curved surface can meet the design requirements.
Owner:HUAZHONG UNIV OF SCI & TECH

Method and system for collecting and processing backlogs based on Java

PendingCN122086365AReduce the number of refreshesImproved front-end docking efficiencyDatabase updatingProgram synchronisationItem CollectionJava
The invention discloses a to-do list collection and processing method and system based on Java, and belongs to the technical field of computer software, collection and processing of to-do lists are achieved through feature dictionary matching, and the method comprises the steps that multi-source data are collected, a to-do feature dictionary is constructed, and to-do information is recognized through a multi-dimensional matching algorithm; the rear end constructs a layered architecture based on the SpringBoot and the SpringMVC, and the SpringMVC is used for providing a standard RESTfulAPI (Application Program Interface) and supporting to-do query and state updating functions; the front end adopts Vue3 + ElementPlus to realize a response type Web interface, and a back-end interface is called through Axios to realize efficient collaboration of the front end and the back end. According to the method, automatic access of multi-source heterogeneous data and extraction of the information to be handled are realized, the accuracy of identification of the backlogs is improved, and automatic classification, priority ranking, reminding triggering and state updating of the backlogs are realized.
Owner:INSPUR SOFTWARE CO LTD

Software testing device for escalator printing plate

The invention discloses a software testing device for an escalator printing plate. The software testing device comprises a relay output module, an IO input board card, an IO output board card, a pulse output board card, a CAN communication board card, an RS232 communication board card, a relay and a testing system. The test system is connected with the relay output module, the IO input board card, the IO output board card, the pulse output board card, the CAN communication board card and the RS232 communication board card. The IO input board card is connected with the relay, and a level output signal of an escalator printing plate is transferred by the relay and then is input into the IO input board card; the IO output board card and the relay output module are connected with the relay, and level output signals of the IO output board card and the relay output module are transferred through the relay and then input into an escalator printing plate level input interface. And the pulse output board card is connected with a pulse input interface of the escalator printing plate. All external interfaces of an escalator hardware system can be replaced, more functions and scenes are covered, and the automation degree of software testing is improved.
Owner:SHANGHAI MITSUBISHI ELEVATOR CO LTD

Approximate izhikevich neuron model and calculation method based on adaptive cordic algorithm

This invention provides an approximate Izhikevich neuron model and calculation method based on an adaptive Cordic algorithm, belonging to the field of neuron model technology. The method includes: obtaining the current residual angle to determine residual angle update parameters and membrane electrical displacement; updating the current residual angle based on the residual angle update parameters to obtain the target residual angle, and continuing to iterate on the target residual angle if it does not meet a preset accuracy; obtaining the membrane potential input, and calculating the membrane electrical displacement value based on the membrane potential input and the membrane electrical displacement; accumulating the membrane electrical displacement values ​​after multiple iterations to obtain the target voltage value, and outputting the target voltage value when the target residual angle meets the preset accuracy. The above method dynamically adjusts the iteration step size according to the residual angle size, quickly approximating the target value to reduce the number of iterations, ensuring the accuracy, reliability, and precision of the final output. The process of residual angle update and shift operation is parallelized, improving the efficiency of iteration.
Owner:INST OF MICROELECTRONICS CHINESE ACAD OF SCI LTD

Wind tunnel model path planning method based on multiple motion mechanisms

The invention discloses a wind tunnel model path planning method based on multiple motion mechanisms, is used for designing operation paths of an airplane model and an external store model in a grid force measurement test of the multiple motion mechanisms, realizes that the airplane model and the external store model efficiently and safely complete all grid points, and belongs to the technical field of grid force measurement wind tunnel tests. Comprising the following steps: (1) establishing a motion simulation model of a motion mechanism, a test model and a wind tunnel; (2) designing a plurality of paths with different motion sequences according to the distribution of test grid points of each test model; (3) setting the movement mode of the movement mechanism to be simultaneous movement at the maximum running speed, then carrying out analogue simulation on a plurality of paths, selecting a path A and a path B which consume less time in the first two as to-be-selected paths, and assuming that the path A is the path which consumes the shortest time; and (4) carrying out collision detection simulation on the path A and the path B, and determining a final test implementation path.
Owner:CHINA ACAD OF AEROSPACE AERODYNAMICS

A software code optimization method and system based on big data

PendingCN122285475AImprove optimization accuracyIncrease adoption rateObject codeTheoretical computer science
This invention discloses a software code optimization method and system based on big data. The method includes the following steps: constructing a code optimization knowledge base; acquiring the target code to be optimized and its corresponding estimated runtime data features; parsing the target code, extracting its structural features, and fusing the structural features with the estimated runtime data features to generate a target code fingerprint; performing similarity matching between the target code fingerprint and historical code fingerprints in the code optimization knowledge base to determine target historical code fingerprints that match the target code fingerprint; and determining a target optimization strategy for the target code based on the optimization strategy corresponding to the target historical code fingerprint. This invention can improve development efficiency, reducing the performance tuning work of senior architects, which can take hours or even days, to automated recommendations at the minute level, greatly improving software iteration efficiency.
Owner:SICHUAN MATIAOSHI NETWORK TECHNOLOGY CO LTD

Static time sequence analysis method, system, equipment, medium and product

The invention provides a static time sequence analysis method, which can be applied to the field of time sequence prediction, and comprises the following steps: obtaining back-end item iteration information and a configuration file of an SOC sub-module to obtain configuration information, the configuration information comprising a working mode; according to the configuration information, a time sequence analysis tool is called to extract and integrate static time sequence analysis data files of the SOC sub-modules, and a time sequence data set corresponding to the working mode is generated; and calling the deep neural network corresponding to the configuration information based on the time sequence data set corresponding to the working mode, and outputting an analysis report of the SOC sub-module. According to the method, the configuration information is closely matched with back-end project iteration, the corresponding deep neural network is called, analysis reports in different working modes are automatically output, static time sequence analysis of the SOC sub-module is accelerated, the total time of static time sequence analysis in project design is shortened, and the iteration efficiency is improved. In addition, the invention also provides a static time sequence analysis system, electronic equipment, a computer readable storage medium and a computer program product.
Owner:PICOCOM (HANGZHOU) CO LTD

Control method of vehicle, computer device and storage medium

The present disclosure relates to a vehicle control method, device, computer equipment, storage medium and computer program product. The method comprises: obtaining to-be-predicted vehicle data of a vehicle; inputting the to-be-predicted vehicle data into a target perception model, and outputting a perception prediction result through the target perception model, wherein the target perception model is obtained after an initial perception model is optimized based on labeled vehicle data samples and is verified to pass, the labeled vehicle data samples include vehicle data samples labeled with a perception result label, the vehicle data samples are obtained when an initial perception result of the initial perception model has an error, and the perception result label has a correlation with a problem type corresponding to the error; and controlling the vehicle in a control mode matched with the perception prediction result. The present method can improve the accuracy of vehicle control.
Owner:安徽蔚来智驾科技有限公司

Polishing head and polishing device

The invention relates to the field of polishing machine tools, and discloses a polishing head and a polishing device.The polishing head comprises a main rod, a second rod, a third rod and a fourth rod, the piezoelectric actuator is connected into the mounting groove, the piezoelectric actuator is provided with a moving end capable of moving up and down, and the moving end is arranged downwards; the transmission part is connected to the moving end and downwards extends out of the mounting groove; the piezoelectric actuator is used for performing micro-motion control on the polishing disc, the movement of the polishing disc can be quickly and accurately controlled, so that the machining efficiency and quality of the polishing head are improved, the polishing disc can be well attached to the surface of a workpiece in the polishing machining process for the surface of the workpiece which is not in rotational symmetry, and the machining precision of the polishing head is improved. And in combination with a smooth track and dwell time controlled by a computer, the surface shape iteration efficiency is improved, so that the machining universality of the polishing head on different workpieces is further improved.
Owner:JIHUA LAB

Ground disaster high-risk area landslide cooperative monitoring and early warning method and system

The invention discloses a cooperative monitoring and early warning method and system for landslide in a high-risk area of ground disaster, and is applied to the technical field of intelligent monitoring and early warning. The method comprises the following steps: acquiring landslide surface data and landslide geologic structure data of a target landslide; setting a plurality of monitoring devices; acquiring landslide motion type data and landslide deformation type data; obtaining predicted displacement data; calculating a first correction weight, and correcting the first correction weight to form a second correction weight; updating the second neural network model; and judging whether the target landslide is safe or not according to the predicted displacement data output by the updated second neural network model. According to the method, different landslide motion types and landslide deformation types are analyzed, and the analyzed result is applied to the updating process of the neural network model, so that the model prediction precision is effectively improved, the iteration efficiency is improved, the prediction model can be updated to the latest state in time after various emergency situations, and the prediction efficiency is improved. And the safety and the applicability are effectively improved.
Owner:SICHUAN SHUTONG GEOTECHNICAL ENG CO

Vehicle-mounted communication matrix splitting and automatic distribution method, system, device and medium

PendingCN122679171AImprove iteration efficiencyAutomate delivery of results
This invention provides a method, system, device, and medium for splitting and automatically distributing in-vehicle communication matrices, belonging to the field of vehicle technology. The method includes: acquiring a target vehicle communication matrix; acquiring a network segment and node configuration table and a gateway node configuration table; acquiring custom configuration splitting rules and determining splitting dimensions based on these rules; performing multi-dimensional splitting of the target vehicle communication matrix based on the splitting dimensions and the network segment and node configuration table and gateway node configuration table, obtaining split sub-matrix files; and converting the split sub-matrix files into target format files and distributing them. This invention improves iteration efficiency by dynamically splitting according to custom configuration splitting rules and configuration tables, automatically converting multiple formats, intelligently distributing based on roles, and supporting incremental splitting based on change history, thereby achieving automated delivery of results from the final version of the target vehicle communication matrix to various ECU suppliers.
Owner:DONGFENG MOTOR GRP

MBSE-based all-electric chassis design modeling method and device

The invention provides an MBSE-based full-electric chassis design modeling method and device, and solves the technical problems that the design work energy consumption is high in demand tracing and influence analysis, and the simulation calculation of each specialty is dispersed and cannot be integrated and verified. The method comprises the following steps: a demand quantification process: quantifying an overall demand to form a demand model; an architecture forming process: designing functional compositions and physical compositions according to the demand model, sequentially forming organizational structures of a subsystem, a communication architecture and a power supply and distribution architecture, and forming organizational structure optimization according to the demand model; in the demand parameterization process, demand parameterization of the population and the subsystems is formed according to the demand model and the organization structure, and parameter attributes of the population and the subsystems are optimized through inter-system joint simulation; and in the scheme comparison process, multi-scheme comparison is established according to parameter constraints between the overall system and the subsystems, and a UI is provided for data display of the comparison schemes. The method is suitable for design and development of a vehicle chassis scheme, and has the characteristics of high efficiency, universality and reusability.
Owner:BEIJING INST OF SPECIALIZED MACHINERY

Human body use state dynamic evaluation method and system based on artificial intelligence

The invention discloses a human body use state dynamic evaluation method based on artificial intelligence. The method comprises the steps that state information used for reflecting the overall human body use state is acquired; analyzing the state information based on an artificial intelligence model to judge whether the human body use state is suitable for continuing or not; outputting a judgment conclusion according to an analysis result, wherein the judgment conclusion is used for indicating whether the human body use state is allowed to continue; in the continuous use process of the human body, the steps are periodically repeated to keep the consistency and stability of judgment. The invention further discloses a human body use state dynamic evaluation system based on artificial intelligence. By using the method or the system, the overall state in the continuous use process of the human body can be dynamically tracked and evaluated, the long-term consistency and stability of judgment are ensured, and running state judgment parameters of the human body use state lacked in an existing diagnosis system are complemented; the judgment result has a blocking type control characteristic; and the judgment layer and the execution scheme are structurally decoupled.
Owner:王杨

Method and device for recovering polarization of deep medium based on feedback of polarized photoacoustic signal

ActiveCN122546462BImplement proactive recoveryStrong deep polarization recovery ability
The present application relates to the technical field of optical regulation, in particular to a deep medium polarization recovery method and device based on polarization photoacoustic signal feedback, comprising: pulsed laser is incident into a first SLM, and the absolute phase of light is updated by photoacoustic signal to realize the regulation of absolute phase; a half-wave plate, a second SLM and a quarter-wave plate constitute a polarization regulation channel, and the polarization state of light is updated by linear polarization degree to realize the polarization regulation of relative phase; and the additional absolute phase caused in polarization regulation is calculated and reversely applied on the first SLM for compensation, and finally the polarization regulation is realized. The method proposed in the present application overcomes the limitation that the traditional optical imaging is difficult to realize the simultaneous regulation of phase-polarization, and through accurate phase-polarization regulation, different feedback signals are continuously iterated to realize the light energy enhancement and polarization state recovery of deep scattering medium, which can be applied to the polarization detection field of biomedicine.
Owner:SOUTH CHINA NORMAL UNIV +1

Project resource collaborative planning method based on double-triangle evolution mechanism

The invention discloses a project resource collaborative planning method based on a double-triangle evolution mechanism, relates to the field of project resource scheduling, and aims to solve the problem of task execution sorting and project resource scheduling for a complex task network and multiple types of complex resources in a research and development project. The algorithm efficiency is optimized, and the resource utilization rate and project completion quality are improved. For the problem, the invention provides a method based on a double-triangle mechanism, and the method covers the following aspects: abstracting characteristics and models of double-triangle evolution by integrating factors such as a resource supplement mechanism of the double-triangle mechanism, resource scheduling task coding of research and development type scientific research projects, failed latitude processing and the like. Based on the model, an algorithm is provided to process the project resource scheduling problem. The algorithm design comprises individual elimination and supplement based on a double-triangle mechanism, failed latitude processing and research and development type scientific research project resource scheduling task coding.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

A method for obtaining the convection heat transfer coefficient of solid three-dimensional thermal conductivity

The present invention relates to a method for obtaining the convective heat transfer coefficient of solid three-dimensional thermal conductivity, comprising the following steps: placing a solid to be tested in a uniformly flowing fluid, obtaining the fluid temperature, the initial surface temperature of the solid to be tested, the thermal conductivity and the thermal diffusion coefficient, measuring a certain temperature reached by the surface of the solid to be tested and the time consumed to reach the temperature, and recording them as the measured temperature and the measured time respectively; calculating the surface temperature of the solid to be tested at a certain moment within the measurement time, constructing a temperature cost function and a Lagrangian function for measuring the convective heat transfer coefficient; estimating the initial convective heat transfer coefficient, respectively obtaining the temperature distribution and the adjoint operator of the solid surface at all moments within the measurement time, and obtaining the gradient of the convective heat transfer coefficient change; obtaining the optimal convective heat transfer coefficient by the Newton iteration method. Compared with the prior art, the present invention can accurately obtain the heat transfer coefficient considering the influence of the three-dimensional thermal conductivity of the tested firmware, and has no restrictions on the shape of the surface of the solid to be tested.
Owner:SHANGHAI INTELLIGENT & CONNECTED VEHICLE R & D CENTER CO LTD +1

Multi-objective optimization apparatus based on dynamic elitist learning and dynamic environment selection

This invention provides a multi-objective optimization device based on dynamic elite learning and dynamic environment selection, comprising: an input module for user input of m objective functions and n decision variable bounding range functions constructed according to a specified multi-objective optimization task; an initialization module for constructing and initializing a population P consisting of N individuals, and setting the iteration count G to 1; a population storage module for storing the population P; a counting storage module for storing the iteration count G and a pre-set maximum number of iterations; and a population iteration module for iteratively updating the population P based on the iteration count G, the maximum number of iterations, the decision variable bounding range functions, and the objective functions. F The output module is used to display the population P to the user. F As a preferred solution for multi-objective optimization tasks, this method can more quickly and accurately obtain the preferred solution for a given multi-objective task.
Owner:TONGJI UNIV

A target detection method, apparatus and electronic device

This application provides a target detection method, apparatus, and electronic device. This embodiment uses weakly supervised learning to determine the labels of target candidate regions containing target objects. Weakly supervised learning does not require manual, precise labeling of the type corresponding to each sample; it only needs to record the user's operation information on the target candidate region. This process does not involve manual labeling of the target candidate region, thereby reducing the labor and time costs of labeling and improving the iterative efficiency of the detection model.
Owner:HANGZHOU HIKVISION DIGITAL TECHNOLOGY CO LTD

Gameplay parameter management method, device, system, medium and computer program product

The present disclosure relates to a gameplay parameter management method, device, system, medium and computer program product. In at least one embodiment of the present disclosure, by providing a gameplay parameter editing page, it is convenient for business personnel to directly edit the gameplay parameters on the page, and then generate a gameplay parameter form to achieve business personalized customization of the gameplay parameters; after generating the gameplay parameter form, the business personnel can bind the gameplay parameter form with the activity configuration item; by providing an activity editing page, the various activity configuration items involved in an activity are organized into an activity configuration item list, which is convenient for business personnel to assign parameters to the gameplay parameter form bound to each activity configuration item, and then generate an activity, thereby achieving management of gameplay parameters in units of activities, and business personnel can directly participate in management, avoiding the problem of insufficient number of R&D personnel or long development cycle, and not affecting the efficiency of business iteration.
Owner:HANGZHOU ALIBABA INT INTERNET IND CO LTD

Image testing method, device, computer equipment and storage medium

The present application relates to an image testing method, device, computer equipment and storage medium. The method comprises: obtaining at least one benchmark test atlas; wherein each benchmark test atlas comprises multiple task atlases, and each task atlas comprises multiple base library fingerprint maps and / or multiple unlocking fingerprint maps; obtaining the task size of each task atlas; running the target number of target processes according to the number of task atlases and the maximum number of processes allowed to run; according to the task size of each task atlas, assigning each task atlas as a task to be tested to the corresponding target process according to a preset allocation strategy and performing a running operation to obtain a test record; wherein the running operation comprises an entry operation for multiple base library fingerprint maps and / or an unlocking operation for multiple unlocking fingerprint maps. The use of this method can improve the efficiency of image testing.
Owner:BEIJING JIIOV TECH CO LTD

Virtual model rendering method and device, storage medium and electronic device

ActiveCN115375813BImprove iteration efficiencySolve technical problems with low iteration efficiency
The application discloses a virtual model rendering method and device, a storage medium and an electronic device. The method comprises the following steps: obtaining an art resource of a model to be rendered; dividing the art resource to obtain model information, material information and association information, wherein the model information is used to represent attribute information of the model to be rendered, the material information is used to represent material performance of the model to be rendered, and the association information at least comprises a corresponding relationship between the model information and the material information; generating a plurality of material layers based on the material information, wherein each material layer corresponds to different types of materials; and rendering the model to be rendered based on the plurality of material layers, the model information and the association information to obtain a first target model. The application solves the technical problem that, in the prior art, due to the mutual coupling of materials and models, the material iteration efficiency is low when the virtual model is rendered.
Owner:NETEASE (HANGZHOU) NETWORK CO LTD