31 results about "Python (programming language)" patented technology

The invention provides an artificial intelligence model standardized training platform and an automatic system. The training platform comprises a training platform body and an automation system. The method is developed on the basis of Python (programming language). A Docker (application container engine) virtualization technology is deployed to a server with a GPU (graphics processing unit). A TCP / IP (Transmission Control Protocol / Internet Protocol) is adopted for communication, is applied to the field of machine learning, and supports frameworks such as mainstream deep learning (DL) TensorFlow, PyTorch, Caffe, Keras, Thheano and the like; a user only needs to click a button or use some script commands on a client, codes can be automatically uploaded, data is downloaded, a GPU (graphics processing unit) is used for training a model of the user, and model evaluation and automatic optimization iteration are completed. In the whole process of training the model, the invention provides a set of operation system to ensure the automation, standardization and specialization of the training process of the artificial intelligence model, save time and simplify steps.

The invention belongs to the technical field of natural language processing, and particularly relates to a method for automatically generating Python codes from natural languages. The method comprisesthe following steps of: 1, generating an abstract syntax tree of a program fragment according to natural language description by adopting a generator of the GAN; 2, judging whether the semanteme of the abstract syntax tree generated by the generator is consistent with the semanteme described by a given natural language or not by adopting a GAN discriminator; and 3, training a generator and a discriminator of the GAN together. According to the method, a code generation system is generated through generative adversarial network optimization training, and the system can generate a section of program code with the same function according to natural language description of one function given by a user. Compared with a traditional optimization method, the generative adversarial network is usedfor adversarial game training, and the generator can learn language models of natural languages and programming languages more effectively.

The invention discloses a carburizing hardened gear contact fatigue risk assessment method with residual stress being considered. The method comprises the following steps: 1) establishing a two-dimensional contact mode through an ABAQUS platform; 2) testing distribution curve of residual stress and distribution curve of hardness of a gear sample; 3) according to the hardness curve of the gear sample and linear relation between the hardness and yield strength, obtaining a distribution curve of partial yield strength of the gear in the depth direction; 4) adding material attributes and an initial residual stress value to a finite element model through a Python programming language and combined with material local yield strength distribution curve and residual stress distribution curve; 5) through Dang Van multi-axis stress criterion, calculating equivalent stress of subsurface of the gear; and 6) calculating a fatigue failure risk value at any point in a key contact area. The technical effects are that, the method estimates contact fatigue risk of the carburizing hardened gear under the condition of considering the residual stress, thereby reducing loss of production benefits causedby contact fatigue failure of the hardened gear.

The invention discloses a finite element modeling method for a variable stiffness composite laminate. The method includes performing programming through Python and a PCL language; directly extractingthe fiber angle and node information of a laminated plate shell unit in a finite element model established by a Patran finite element through an operation file; calculating a central node coordinate of each shell unit, if the size of the shell unit is relatively small, approximately regarding the fiber angle of the point as the fiber angle of the unit, and finally re-endowing the fiber angle of each shell unit with the attribute of the shell unit according to a set curve track, thereby realizing fiber curve laying of the laminated plate. A variable stiffness laminate model is obtained by combining Patran and Python, then the model is submitted to Nastran to be solved and analyzed, and visualization processing is carried out through Patran import. According to the method, two programming languages of Python and PCL are combined, other platforms are not needed, and finite element modeling of the variable-rigidity laminated plate can be achieved through Patran.

The invention relates to a smart energy system visual optimization design and operation integration device and method. The device comprises a data acquisition module which acquires a design file or real-time operation test data; a data preprocessing module which preprocesses the acquired data through the Python and other programming language; an optimization module which establishes a smart energysystem simulation model based on the preprocessed data and applies an optimization engine on the simulation model so as to acquire the optimization data; and a post-processing module which post-processes the optimization data through the Python and other programming language, provides an optimization design scheme in optimization design according to the post-processing file and applies the optimization design scheme to system design configuration, provides an optimization operation strategy in optimization operation according to the post-processing file and applies the optimization operationstrategy to system operation and outputs the final optimal value after continuous adjustment and optimization and displays the final optimal value. Compared with the devices and the methods in the prior art, optimization design and optimization operation are combined together so as to have the advantages of convenient operation and high accuracy.

The invention relates to an automatic chip register development system based on template variable replacement. The automatic chip register development system comprises an input file; the input file comprises a register document file, an RTL code, UVM verification, a control parameter file in C and C + + formats, and a template file in an output file format; in a register document file, registers are subjected to level-to-level management, and three levels including a chip level, a module level and a register group are included; according to the method, a Python programming language is used for generating a Python class for each register, a register group and a module level, namely a chip level. Meanwhile, code generation control parameters are converted into the Python classes, then different template files are called, register variables in template level files are replaced according to control parameter values, and finally design, verification and debugging codes of the chip registers are automatically generated.

The invention discloses a northern plain area large-scale rural residence layout generation design method, and belongs to the field of digital urban and rural planning. According to the method, three-level generation rules of rural residential base combinations in northern plain areas are determined; wherein the first-level rule describes a residence unit mode of four-household combination, the second-level rule describes a public space organization, a road network structure and a neighborhood mode of every 16-household combination, and the third-level rule describes land building density, land inside and outside connection and a community mode of every 64-household combination. A homestead combination prototype of each level of rule is built through a Python programming language, samplesgenerated by each level of rule are automatically generated and screened, and a corpus in an npy format is built to store sample information; for a dwg file of a given land parcel, base information isidentified by means of a PyautoCAD module, and a sample is automatically called from a corresponding sample pre-stock library to be generated and designed. Automatic design of large-scale rural residence layout in northern plain areas is realized.

The invention discloses a system for constructing a complex sql statement by utilizing python syntax semantics, and belongs to the technical field of script programming and structured query language.The system comprises a decomposition unit, an assignment unit and a series unit. The decomposition unit is used for decomposing the complex sql query statement to obtain clauses in various forms; theassignment unit is used for assigning the clauses in various forms obtained by decomposition to a python variable; and the serial connection unit is used for serially connecting and combining the clauses in various forms obtained by decomposition together according to the assigned python variable to generate a complex sql statement. According to the system for constructing the complex sql statement by utilizing the python syntax semantics, the flexibility of logic processing of a programming language is utilized, the original complex logic is split and processed, the complex problem is simplified, and compared with the prior art, the readability of a program is improved, the later expandability is also improved, and the maintenance cost is reduced.

The invention discloses a dispatching operation data automatic acquisition and submission management system, which realizes automatic login and acquisition of power grid operation data webpages among different systems through a python programming language under the condition that an internal network of a power system is closed and the network security is low in computer user authority. The scheme comprises the following steps: capturing related historical data at a specified position of a computer, summarizing power grid operation data based on a set power grid report rule, and processing the extracted power grid operation data on the crawled information by using Beautiful Soup; performing interactive operation on the data in the summary report and historical data in a historical database according to various set logics; judging reasonability of data and automatically reporting data to a related system after checking data to be correct. The system solves the problems that a dispatcher needs to log in different systems to copy and collect power dispatching data and organize the power dispatching data into reports to be reported to provincial middle dispatching, data sources are diversified, data units and data generation time of various systems are inconsistent, data arrangement wastes time and manpower, the accuracy is not high, data processing methods are lacked are solved.

The invention discloses a method for converting a plaintext of a data table into a binary ciphertext convenient for a program to read, and the method provided by the invention needs to be combined with an open-source Protobuf (version 3.11. X) provided by Google, and is based on a cross-platform programming language Python. Protobuf is an open-source data definition format provided by Google, each column of a data table can be mapped to a specified field in a data structure defined in Protobuf by using Protobuf, a converted binary file is read in a program and converted into an array of a data structure type corresponding to Protobuf, the array is operated, the data hidden in the data table such as the primary key, the maximum value and the minimum value can be packaged, data access can be carried out according to the row number and the column number without reading the Excel file, access can be directly carried out through members of the mapped data structure, and the code readability and the development efficiency are improved.

The invention discloses a programming language Python-based database operation method and system, and the method comprises the steps: obtaining a service operation request which is used for requesting the operation processing of target data of a to-be-accessed database; and calling a target operation function pre-configured for the to-be-accessed database in a database operation system, and processing the service operation request so as to perform operation processing of the target data in the to-be-accessed database. By means of the method and device, the problems that in the prior art, a database operation process is complex, operation is not easy and convenient, and performance is low can be solved.

The invention provides an automatic generation method of a programming problem post title based on Transform, and belongs to the technical field of computer application. According to the technical scheme, the method comprises the following steps of: (1) collecting question posts related to four programming languages, namely Java, C #, Python and JavaScript, on a question and answer website of a developer, and filtering out low-quality question posts on the basis of three heuristic rules; and (2) on the basis of the constructed corpus, performing fine tuning on the constructed T5 model based on Transform, and performing optimization by using an AdaFactor optimizer and an attention mechanism to construct an automatic question post title generation model. The title generation method has the beneficial effects that a simple and accurate title is generated for a question post of a question and answer website (such as Stack Overflow) of a developer, and automation of question post title generation for four programming languages of Java, C #, Python and JavaScript is realized.

The invention provides a method for remotely deploying an MATLAB function. Compared with a method for remotely deploying an MATLAB Web Server, the method for remotely deploying the MATLAB function has the advantage that the online access speed is obviously improved. The method comprises the following steps: designing a front-end UI interface by using a front-end programming language, and carrying out data interaction with a Python rear end through a progressive front-end framework Vue; the method comprises the following steps of: constructing a back-end API (Application Program Interface) by using an asynchronous Web framework FastAPI (Application Program Interface) in the Python, taking Uvicron as a server, enabling the back end of the Python to call an MATLAB function for calculation by using a third-party packet MATLAB Engine For Python in the Python, and returning a result to a front end; in the Python, a Jinja2 template rendering engine is used for realizing front-end page rendering; an Nginx server is used as a picture server, so that an extranet user can see a picture file in a calculation result in a front-end interface after calculation. According to the method, the MATLAB function is remotely deployed in another quicker mode, the access speed at a user side is high, early-stage development is simple and easy to master, and later-stage maintenance is simple and convenient.

The invention provides a novel low-energy-consumption low-power embedded intelligent camera system. The system mainly comprises a hardware integrating device, a software integrating device and a database. A sensing module, a control module and a multifunctional module are embedded into a mainboard, the modules are connected together through a control serial bus to realize a transmission function,and components include a microprocessor, a memory, a network system and an interrupter. According to positions and functions of different modules, a Python programming language based application program interface is adopted to control image algorithms and functions in the database to perform corresponding operations on inputted images, and images with application result prompts are outputted directly. An embedded platform based camera shooting processing system can be provided, image shooting, embedding processing and application promoting and packaging are integrated, and low energy consumption is realized by adoption of an interrupt mechanism for operating power reduction.

The invention relates to the technical field of software design, in particular to CFD automatic integration calculation software based on Python. According to the method, the problems that a large amount of software is used and the consumed manpower and time cost is high in conventional fluid machine CFD are considered, the purpose of simplifying the fluid machine CFD is achieved, the CFD automatic integration calculation software designed through the Python programming language is used, multiple pieces of software can be conveniently integrated, full-automatic CFD except design can be achieved, manual participation is greatly reduced, and the manpower and time cost is reduced.

The invention discloses a method for remarkably improving the online access speed of MATLABWebApp. The method comprises the following steps: (1) judging whether the WebApp needs to be subjected to file conversion or not according to the type to which the WebApp belongs; (2) creating an application layout of the MATLABWebApp by using a front-end technology; (3) writing a back-end program by using a high-level programming language Python, finding the MATLAB EngineAPIforPython under an installation directory of the MATLAB, and importing the MATLAB EngineAPIforPython into a library file of the Python; (4) installing and configuring an Nginx server; (5) installing and configuring an NAT tool; and (6) after the web application completes response, the browser of the user respectively receives a numerical data result from the back end and other results from the Nginx end. According to the method, the problem that the MATLAB application is slowly loaded at the Web end is solved, the front-end technology is used for designing the application, the PythonWeb technology is combined for building a deployment platform, the design and deployment of the application are modularized, the two modules perform own functions and supplement each other, and the development quality of the application and the use experience of a user are remarkably improved.

The invention provides a distributed network asset scanning detection platform and a scanning detection method. The platform comprises a distributed network asset scanning detection subsystem constructed based on a Kafka distributed message publishing and subscribing system; the distributed mass data elastic access subsystem is constructed on the basis of an ElasticSearch distributed mass data storage and retrieval system; the Web Server server side is written on the basis of a Flask framework, an elasticsearch 6 framework, a kafka-python framework and a pymysql framework by utilizing a Python programming language, the Web Server server side is connected with the distributed network asset scanning and detecting subsystem, and the distributed network asset scanning and detecting subsystem is connected with the distributed mass data elastic access subsystem; a MySQL database, wherein the Web Server server side is connected with the MySQL database by utilizing a pymysql framework; and the Web page is constructed by utilizing a Vue framework, and the Web page is connected with the Web Server server side. Through the technical scheme of the invention, scanning and detection of network assets are realized, the parallelization degree is high, and the efficiency of processing large-scale scanning and detection tasks is high.

Accessing a schema-free database includes constructing a model indicating a structure for the data to be used by applications accessing the data, validating the model based on the structure and on the data stored in the schema-free database, providing an API based on the structure, and accessing the database using the API. The model may be constructed by extracting data structure information from a program. The program may be written in the Ruby programming language or the Python programming language. The API may be a RESTful API.

The invention provides a multi-language geoscience online service method and device, a storage medium and electronic equipment. The geoscience online service method based on Docker encapsulation is realized. And through online resource scheduling, task scheduling such as preparation of online algorithm input data, algorithm operation and ending, post-processing of output data and the like is realized. A user operates at a webpage end, a calculation instruction is sent to the online algorithm, the online algorithm carries out online calculation according to the calculation instruction, and finally a calculation result file is displayed in a webpage. Meanwhile, multi-language geoscience online services are achieved, the method can be suitable for various programming languages, operation environments such as a C language compiler, a Fortran compiler, a Python compiler and other dependent software programs can be conveniently configured, the algorithm operation result is visually processed, vector and grid space data are displayed in a map visualization mode, and the long-time sequence data is displayed in a chart visualization mode.

The invention discloses a vibration energy transfer evaluation method for a multi-cell structure sole. The method comprises the following steps: S1, establishing an original sole three-dimensional entity model; s2, constructing multi-cell structures of different lattice types, and respectively filling the multi-cell structures into the heel areas of the soles to obtain multi-cell structure soles of different lattice types; s3, constructing a finite element model of the multi-cell structure shoe sole in the step S2; s4, performing steady-state dynamic analysis on the multi-cell structure sole to obtain a response condition of the multi-cell structure sole to vibration; s5, calculating and drawing an equivalent mechanical admittance distribution cloud picture of the multi-cell structure shoe sole by utilizing a Python programming language, and collecting the distribution condition of the equivalent mechanical admittance of different areas of the multi-cell structure shoe sole; and step S6, based on the distribution condition of the equivalent mechanical admittance of the different areas of the multi-cell structure sole, calculating and collecting the distribution condition of the equivalent vibration transmissibility of the different areas of the multi-cell structure sole by using a Python programming language.

The invention belongs to the technical field of computer programming languages and application, and discloses an explanation system and method for Python programming language XD-M which comprises the following steps: for basic statements in an XD-M language and statements from a modeling simulation verification language MSVL, carrying out interpretation by a method of calling a bottom layer interface of an MSVL interpreter; for statements with XD-M language characteristics, creating an equivalent MSVL syntax tree for the statements, expanding an underlying interface, and writting an interpretation method for the statements by an underlying layer. An XD-M program written by using the XD-M language can be subjected to modeling, simulation and verification, so that the correctness, reliability and safety of the XD-M language are improved, a Python-like simple variable is used without a programming style of type declaration, and the flexibility of the XD-M program is improved; the invention depends on a MinGW development environment.

The invention discloses a large-scale rural residential layout generation and design method in the northern plain area, which belongs to the field of digital urban and rural planning. The present invention determines the three-level generation rules for the combination of rural homesteads in the northern plains: the first-level rule describes the residential unit pattern of the four-household combination, and the second-level rule describes the public space organization, road network structure, and neighborhood pattern for every 16-household combination , the third-level rules describe the building density of the land, the internal and external connections of the land, and the community model of every 64 household combinations; and establish the prototype of the homestead combination of the rules at all levels through the Python programming language, automatically generate and screen the samples generated by the rules at all levels, and establish The corpus in npy format is used to store sample information; for the dwg file of a given plot, the PyautoCAD module is used to identify the base information and automatically call the sample from the corresponding sample prediction library for design. The invention realizes the automatic layout design of large-scale rural houses in the northern plain area.

The invention discloses a steel ball surface defect detection method based on a single stage, and compared with traditional manual detection, the detection method has higher detection precision, detection efficiency and robustness. The single-stage detection method is based on a YOLOv4 network structure pre-training model. The method mainly comprises the following steps of steel ball surface image data acquisition, image augmentation, data set image annotation, data set division, pre-training model construction, model training and model verification. The model of the method can automatically extract the characteristics of the surface defects of the steel ball, and can accurately and quickly detect the positions of various defects on the surface of the steel ball. According to the method, a Python programming language is used, a Keras framework is used as a front-end function to realize, Tensorflow is used as a rear-end data processing, and GPU (NVIDIA, GTX1080Ti) is used for model training, model verification and model testing to obtain corresponding evaluation indexes and test results. According to experimental results, the method can be used for quickly and accurately detecting the surface defects of the steel balls.

The invention discloses a method for managing network equipment based on an NETCONF proxy server, which comprises the following steps of: firstly, developing the NETCONF proxy server by utilizing a software programming language such as Python; and then the NETCONF proxy server intervenes in the RESTCONF server of the network equipment, the XML transmission format of the NETCONF is converted into the JSON transmission format of the RESTCONF, and finally the purpose of managing the network equipment through the RESTCONF server is achieved. The method not only ensures that the network equipment only needs to maintain one set of RESTCONF server, but also supports two sets of RESTCONF / NETCONF network equipment management schemes, thereby greatly reducing the complexity of the equipment server and improving the stability of the system. The method provided by the invention has strong practical application significance.

The invention discloses a controllable customized back-end development system and method, and relates to the technical field of code development, the controllable customized back-end development system comprises a graphical service orchestration tool and a translation engine, the graphical service orchestration tool is used for providing graphical service orchestration with complete Turing, and the translation engine is used for translating the graphical service orchestration. And the translation engine is used for translating and generating a runnable programming source code according to a group of primitives designed by a user. According to the controllable customized back-end development system and method, complex code programming is replaced with simple graphical service arrangement, after a user carries out service arrangement through a graphical tool, the graphical service arrangement result can be translated into various programming languages of Java, Python and JavaScript according to the logic relation, defined by the user, of primitives and connecting lines in a graph, and therefore the user can carry out service arrangement through the graphical tool, the graphical service arrangement result can be converted into the multiple programming languages of Java, Python and JavaScript. An operable source code of a specified programming language is directly generated, and then a new arranged service interface is formed through source code compiling.