90 results about "Training material" patented technology

The invention relates to an image migration method based on neural rendering and depth priori, and the method comprises the steps: 1, obtaining a group of discrete multi-view images of the same object in a static scene, the camera coordinates and orientation of the corresponding images, and a target material map; 2, building an object three-dimensional reconstruction network; 3, building a generative adversarial network for training material migration; 4, training the three-dimensional reconstruction network in the step 2 by using the training data set processed in the step 1, and training the material migration network in the step 3 based on the output of the network in the step 2, the training data set in the step 1 and various loss functions until the network converges; step 5, for an image to be migrated, carrying out migration processing by using the image migration model trained in the step 4 to obtain a final migrated image; the migration image generated by the method has high-quality material details, and the problems that an existing image migration method is blurred, multiple view angles are inconsistent and the like can be effectively solved.

An AI-driven emotion recognition system for learning adaptation in educational institutions, comprehensive: • an emotion recognition unit that tracks students' facial expressions in real time and categorizes the resulting expressions into a specific emotional class from a given set using a Convolutional Neural Network (CNN) trained on a large emotion dataset; • a real-time processing unit that extracts facial features and dynamic emotional characteristics of students through adaptive filter algorithms; • an LMS integration unit, adapted to combine discovered emotional knowledge with online and offline learning environments for customizable learning adjustments; • a personalization engine programmed to intelligently adapt training content, response approaches, and training materials according to the identified emotional states, so that each learner receives individually tailored and more intensive learning experiences; • a feedback loop that sends educators real-time alerts and recommended videos to address emotional feedback or lesson gaps, allowing teachers to tackle negative signals in the moment (e.g., frustrated face, etc.); • AI data storage and adaptive learning model unit that stores anonymized data for emotion recognition, updates the AI ​​model and improves it through learning over time, and ensures that the use of the information is in compliance with data protection regulations; • A user interface module specifically designed for educators, providing a visual overview of students' emotional patterns and real-time analysis of their emotions and engagement as a basis for important decisions that lead to improvements in teaching processes.

An engineering machinery equipment performance degradation prediction method based on multi-model stacking fusion is characterized in that design calculation, a simulation model and a data driving model are combined, physical test data under different carrying equipment and different working conditions are used as training materials, confidence weight distribution of sub-models is completed, and a fusion total model is obtained. And when real object live test data under certain conditions are lacked, the environmental conditions are used as input, a final prediction value is obtained through a model stacking weight distribution mechanism, and the prediction result of the performance degradation of the engineering mechanical equipment is optimized.

A system for generating instructions based on part from human training materials, wherein the generated instructions are used to train automated agents. The training materials may include training manuals, knowledge-base documents, articles, and other content. The training materials are processed and converted to a format that is digestible by a machine learning model. The processed training materials can be submitted as a prompt, along with role information and prompt instruction information. The prompt instructions provided with the role and training materials may direct the machine learning model to find relevant generated instructions from the provided training materials. The machine learning model processes the prompt and outputs extracted instructions from the content. The instruction document may be stored in a vector database for later use.

The invention discloses a controllable dialogue case generation method based on a large model, a full-process automatic dialogue case processing system is constructed based on a three-stage assembly line of screening, optimization and COT generation, and the method comprises the following specific steps: a first stage: multi-dimensional dialogue screening, including dialogue quality rating model construction and hierarchical screening strategy; in the second stage, multi-role multi-emphasis parallel rewriting is carried out; and a third stage: COT information generation. According to the method, automatic generation of'excellent dialogue cases + COT information 'is achieved, it is ensured that the generated dialogue cases meet the requirements of logic coherence, role consistency and theme focusing, the COT information matched with the dialogue cases is synchronously generated, and standardized training materials are provided for large model learning; a deep correlation mechanism of COT information and case content is constructed, a complete link of'case text-COT reasoning-knowledge support 'is formed, and the learning efficiency of a large model for complex dialogue logic is improved; and the large-scale generation bottleneck of manual dependence is eliminated.

The invention relates to an object material identification method based on knocking sound simulation and deep learning, and the method comprises the steps: carrying out the knocking sound simulation of a given knocking point through a generated model, carrying out the training and comparison through combining with the real knocking sound, and finally carrying out the analysis to obtain the material of a target object. The method comprises two parts of deep learning model training and material identification and classification, and the deep learning model training comprises five steps of training material library construction, object model construction, actual knocking sound recording, knocking sound simulation and model training; the material identification and classification comprises four steps of target object model construction, target actual knocking sound recording, target knocking sound simulation and model identification and classification. Compared with the prior art, the method has the advantages of avoiding collection work of a large amount of actual knocking sounds, being high in material recognition result accuracy and the like.

The invention discloses a short-term memory manipulation ability training method and related products, and the method comprises the steps: determining an initial training level based on the basic information of a target object; determining a material generation condition of current training of the target object based on the initial training level and the basic information; inputting the material generation condition into a large language model to obtain a training material; generating a training task based on the training material and the initial training level; displaying the training task to the target object, and obtaining feedback data of the target object on the training task; and training the target object for multiple times based on the feedback data and the basic information. By using the method, the challenging and interestingness of training are improved, the training effect and efficiency are further improved, and the training participation degree and the user experience are improved.

A question is received at a first language machine learning model. In response to an external machine learning model providing a response to the question and a confidence determination for the response that exceeds a predetermined threshold, the response is injected into the first language machine learning model so that the response overwrites a vector state layer output of the first language machine learning model that provides another response to the question and without modifying original parameters of the first language machine learning model, where the external machine learning model was trained with training material with which the first language machine learning model was not trained.

The invention discloses a personalized safety training content generation method, system and device and a medium. The personalized safety training content generation method comprises the steps that answer data, historical performance data and future operation tasks of operators in safety training are collected; safety knowledge weak points and learning behavior modes of the operating personnel are identified, and related risk association points are determined; based on the safety knowledge weak point, the learning behavior mode and the risk association point, constructing a structured user portrait, based on a deep neural network model fused with an attention mechanism, screening basic training content matched with the user portrait from a dynamic training material library, and adjusting the basic training content to obtain a basic training result; according to the method, the personalized target safety training content is generated, and the safety training is performed on the operator according to the target safety training content, so that the generation efficiency of the personalized safety training content can be improved.

The disclosure describes an expandable construction method for an artificial intelligence-based customer service question and answer system and a computer system using the same. In the expandable construction method for an artificial intelligence-based customer service question and answer, a customer service question and answer text as test material is inputted to test a question and answer artificial intelligence (AI) model. The question and answer AI model automatically outputs a question and answer report based on the customer service question and answer text. An expert artificial intelligence (AI) model automatically outputs an evaluation score based on the question and answer report. It is determined whether to input the question and answer report into the question and answer AI model to serve as training material based on the evaluation score.

Examples related to facilitating the training, assessment, and dispatching of technicians to a customer location for at least the purpose of servicing or repairing machines, devices, or systems present at the customer location are presented. The technician can be selected for dispatch to the customer location by review of database information, including training information for the technician, among other things. While at the customer location, the technician can be provided with training information delivered via a training module, where such training material content can comprise written materials, photos, video, virtual assistance, remote assistance, games, or the like. The technician can generate videos or other collateral while at the customer location, which can be included for use in a library of training materials for subsequent use, or for other purposes. Various systems and processes for facilitating the selection, management and deployment of training materials for technicians are also provided herein.

The invention discloses a mine enterprise knowledge training method based on a large model, and the method comprises the steps: obtaining a question keyword and a question type according to received mine enterprise knowledge question data, obtaining knowledge data according to the question keyword in combination with a knowledge base, and generating a query statement according to the question type in combination with a preset prompt word; according to the queried statements, database query is carried out, and answer texts are obtained through a large model in combination with question keywords and knowledge data; and obtaining a question intention according to the received mine enterprise knowledge question data, and obtaining different forms of training data in combination with the answer text. According to the method, based on deep matching of the question keywords and the knowledge base, accurate queriable statements are generated; generating a multi-dimensional answer text in combination with database service data; furthermore, customized training data are recognized through natural language intentions, and policy adaptation, accurate response and personalized training are achieved.

A system for detecting one or more target materials in an un-calibrated multi-spectral data cube comprising a collection of pixels, the system comprising a processing circuitry configured to: obtain: (A) a machine learning model capable of receiving the un-calibrated multi-spectral data cube and determining for at least one pixel of the pixels at least one material indicator, indicative of existence of a given target material of the target materials at the location of the pixel, wherein the machine learning model is trained utilizing a labeled training-data set comprising of a plurality of training records, each training record comprising: (i) a training un-calibrated multi-spectral data cube, and (ii) at least one training material indicator associated with at least one pixel of the training un-calibrated multi-spectral data cube, indicative of existence of the target material at the location of the pixel, and (B) the un-calibrated multi-spectral data cube; and determine for at least one pixel of the pixels of the un-calibrated multi-spectral data cube, at least one material indicator, and a corresponding calibrated multi-spectral data cube, wherein the corresponding calibrated multi-spectral data cube is calculated by utilizing a calibration process and an atmospheric simulator that simulates a plurality of simulated un-calibrated multi-spectral data cubes by simulation of different atmospheric conditions over the calibrated multi-spectral cube.

The invention provides a content generation method, a related device and a computer program product. The method comprises the following steps: acquiring demand description information of target generated content; performing demand decomposition on the demand description information through a demand processing model to obtain material demand information, training an initial demand processing model according to the difference between the training material and the sample material to obtain the demand processing model, and performing demand decomposition on the sample demand description information by the training material according to the initial demand processing model to obtain the material demand information; after the training material demand information is obtained, the sample material is extracted from a material library based on the training material demand information, and the sample material is used for meeting the material demand of the sample demand description information and is extracted from the material library; extracting a target material corresponding to the material demand information from a material library; and generating target generation content based on the target material. Therefore, the demand processing model with stronger perception ability and stronger material selection ability for the material library can be utilized to more accurately generate the material demand information for obtaining the materials required by the processing demand of the user according to the content in the material library, so that the model illusion is reduced, and the user experience is improved. Therefore, the task of demand processing and content generation can be better completed.

The utility model relates to the technical field of teaching equipment, and specifically discloses a single-person command and training device for virtual-reality combined emergency drills, which includes: a table body, a first display screen component and a second display screen component; a VR device cabin and a mainframe cabin are detachably arranged on the table body; the VR device cabin is used to accommodate the VR device; the mainframe cabin is used to accommodate the mainframe; the first display screen component and the second display screen component are both installed on the tabletop of the table body, and the included angle between the display end faces of the first display screen component and the second display screen component is 90 degrees to 180 degrees; the first display screen component is used to store training materials and perform screen sharing with multiple student terminals; the second display screen component is wirelessly connected to multiple VR devices, and is used to issue VR tasks to each VR device and receive the training data returned by each VR device. The utility model can not only improve the training efficiency of the device, but also improve the layout rationality of each accessory.

The invention belongs to the field of drawing space and legend intelligent recognition, and particularly relates to a method and device for intelligently recognizing CAD drawing space and legend through model training, and the method comprises the following steps: S1, drawing training material manufacturing; importing a drawing training material and a calculation amount drawing; s2, a database and algorithm analysis service is read through DWG; s3, for the data stored in the DWG reading database, on the first aspect, target detection model training is carried out through a view model training data set, and a view model is generated; secondly, data are structured through a drawing; thirdly, importing a legend library; s4, for data of algorithm analysis service, on the first aspect, the algorithm is recognized through a legend; and secondly, identifying an effective space range through a space algorithm. The method greatly improves the efficiency, guarantees the identification precision, reduces the technical threshold, strengthens the cooperation capability, and meets the requirements of multiple scenes.

The invention relates to the technical field of power production safety management, and discloses an intelligent management and control system suitable for power field operation safety. The theoretical basis of the method is that safety operation = creation and maintenance of safety and operation conditions + zero error of people. Comprising a field operation risk analysis module, a safety reminding and checking module, a positioning tracking module, a monitoring analysis module, a training education module and a data management module. A function of automatically acquiring work ticket and work ticket information by associating the work ticket and the work ticket; through operation risk analysis, dangerous point identification and management and control measure functions are realized; safety reminding of an operator is realized through a voice function; an intelligent audio and video system is utilized to monitor an operation site in an omnibearing and dead-corner-free manner, unsafe behaviors of people, unsafe states of equipment and unsafe factors of the environment are accurately recognized, and an audible and visual alarm is given out in time; and a self-learning function is integrated with historical records to continuously perfect an operation dangerous point identification management and control database, and safety training data is perfected.

The utility model provides a workshop equipment safety training device with a storage and placement structure, which comprises a box body assembly and a placement frame, the placement frame is arranged on the side surface of the box body assembly, the box body assembly comprises a storage box, a clamping piece and a training piece, the clamping piece and the training piece are arranged on the upper side surface of the storage box, and the storage box is arranged on the placement frame. Compared with the prior art, the storage box has the following beneficial effects that by arranging the placement frame, when the storage box is used, the three drawers are arranged in the closed space formed by the partition plate I on the right side in the storage box in a sliding manner, so that different types of articles can be conveniently classified and stored; and the placement rack is mounted on the left side surface of the storage box, and a plurality of placement layers are arranged in the placement rack, such a multi-layer structure can make full use of a vertical space, different types of training materials or tools can be placed in a layered manner, and at the moment, a plurality of storage placement structures are formed by matching with an inner accommodating cavity of the storage box, so that the use of a user is facilitated.

The present application relates to a kind of object material identification method based on knocking sound simulation and deep learning, utilize the model generated to carry out the knocking sound simulation of given knock point, then combine real knock sound and train and compare, and finally analyze and obtain the material of target object.The method includes two parts of deep learning model training and material identification and classification, wherein, deep learning model training includes five steps of constructing training material library, object model construction, actual knock sound recording, knocking sound simulation and model training;Material identification and classification include four steps of target object model construction, target actual knock sound recording, target knock sound simulation and model identification and classification.Compared with prior art, the present application has the advantages of avoiding a large number of actual knock sound collection work, high accuracy of material identification result etc..

The invention relates to the technical field of machine learning and block chains, and discloses a laboratory safety education course integration pushing system and method. A trainee grading and classifying subsystem, a course data grading and classifying subsystem and a wrong question grading and classifying subsystem of the system respectively use a deep learning technology to automatically grade and classify trainees, link addresses corresponding to laboratory safety training data, error-prone question types of the trainees and question answering error rates. And the course grading and classification pushing subsystem automatically pushes personalized link address sets corresponding to laboratory safety training data to different trainees by using an unsupervised learning technology, so that the working efficiency is improved. In addition, the course data grading and classifying subsystem integrates link addresses corresponding to laboratory safety training data from different sources inside and outside the school by adopting a block chain technology, and the related problems of communication bottleneck, single-point failure and other privacy leakage caused by a centralized network topology structure can be avoided.

The application discloses a matching layer material parameter prediction method and device, electronic equipment and a storage medium. The sound velocity matrix of the matching layer, the density matrix of the matching layer and preset sound source information are used to obtain sound pressure distribution information in a working field. The sound pressure distribution information is input into an initial matching layer material parameter prediction model to obtain a training material parameter matrix. The initial matching layer material parameter prediction model is updated based on the training material parameter matrix and a filling ratio matrix to obtain a trained matching layer material parameter prediction model. Target skull sound pressure information is input into the trained matching layer material parameter prediction model to obtain a target material parameter matrix. The target material parameter matrix of the matching layer can be obtained based on the target skull sound pressure information. The matching layer made based on the target material parameter matrix can make the sound pressure information in the skull meet the target skull sound pressure information, and the efficiency is high without manual calculation based on experience.

The present application provides a model evolution method, apparatus, electronic device, and machine-readable storage medium. The method includes: determining a training mode for training a target model, where the target model is an algorithm model used by the intelligent analysis device to perform intelligent analysis tasks; the time, resource amount, and / or model training trigger conditions for model training by the intelligent analysis device are different under different training modes; when it is determined that the model training conditions in the training mode are met, training the target model using the obtained training materials according to the training mode; and updating the target model based on the trained target model so that the intelligent analysis device uses the updated target model to perform intelligent analysis tasks. This method can improve the device utilization rate.

The invention relates to the technical field of smart cloud education, and particularly discloses smart cloud classroom practical training equipment which comprises a training table, a training feeding mechanism, a training mechanical arm, a demonstration screen and a control table are installed on the training table, and an electric clamp and a welding gun simulation pen are installed at the execution tail end of the training mechanical arm; after video explanation teaching is carried out through the demonstration screen, training content is displayed and played on the demonstration screen; during practical training operation, an electric clamp on the training mechanical arm is used for clamping a material block, and a welding gun simulation pen is used for welding simulation training; an operation instruction is input through the console, the training feeding mechanism is controlled to convey a training material block to a designated position, and then the operation instruction for controlling the training mechanical arm is input according to the position information of the material block, so that the training mechanical arm runs according to the instruction, and meanwhile, the training mechanical arm can be directly and manually controlled to move; the defects in the prior art are overcome.

The invention provides a method and application for collecting data and amplifying training data of an AI customer service question-answering system, which can collect and train the data of the AI customer service question-answering system, and can organize the collected data to generate a question-answering report. And generating an evaluation score according to the question and answer report to determine whether to input the question and answer report to the AI model as a training textbook. The method for collecting data and amplifying training data of the AI customer service question-answering system comprises the following steps of: receiving initial amplified data; identifying the data type of the initial amplification data, and judging whether the initial amplification data needs to be converted into character type initial amplification data or not according to the data type of the initial amplification data; an AI model outputs a question and answer report based on the character type initial amplification data; an expert AI model outputs an evaluation score based on the question and answer report; and determining whether to input the question and answer report into the AI model as a training textbook based on the evaluation score.

A method for evaluating data to be used to train an object recognition model is to be implemented by a computing device. The computing device stores a plurality of training datasets respectively related to a plurality of images, and each training dataset includes a plurality of entries of training data. The method includes steps of: obtaining, for each image, at least one target area and at least one target property that are related to the image based on the entries of training data; creating, for each image, a training material that includes the image, and the at least one target area and the at least one target property both related to the image; and obtaining at least three object recognition models based on the training materials that are created respectively for the images using at least one machine learning algorithm.

Apparatus and methods for proactively and preemptively communicating with a user interacting with a software application are provided. The apparatus and methods may include an artificial intelligence / machine learning communication engine monitoring and tracking a user's interactions. The apparatus and methods may include the communication engine determining if the user requires further training, if the interaction is fraudulent, and pre-empting requests for information the user may commence. The apparatus and methods may include the communication engine creating and displaying training materials for the user to complete, revoking access if fraud is present, and proactively providing information before the user requests the information.

The invention discloses a training system for security and secrecy, and relates to the technical field of security and secrecy training systems. According to the training system for security and secrecy, the duration of security and secrecy training of an employee in a training period G + 1 is subjected to timing planning by using the assessment score CJ of security and secrecy training of the employee in the training period G and the active learning duration ZD of the employee reading training data between the training period G and the training period G + 1 for active learning; the training duration of each employee is precisely regulated and controlled, the situation that in a traditional technical scheme, all the employees are trained in a unified duration, the training duration is better matched with the actual situation of the employees, the employee training quality can be more effectively improved, the employees can be reminded to actively learn weak courses during training examination, and the training efficiency is improved. The method can help employees to make up for weak items of assessment, reduces the error rate of the employees in daily work, and improves the safety and confidentiality.

The embodiment of the invention discloses an image processing method and device, a storage medium and electronic equipment, and the method comprises the steps: obtaining a customized target IP image picture, carrying out the low-rank fine tuning training of a first diffusion model through employing the target IP image picture, and obtaining a reference IP image generation model, performing training material generation based on the target IP image picture and the reference IP image generation model to obtain at least one IP image generation training material, and performing second diffusion model low-rank fine tuning training on the reference IP image generation model based on the IP image generation training material and the target IP image to obtain a target IP image generation model.

The invention discloses a machine learning-based transformer area line loss acquisition calculation factor anomaly diagnosis method, which comprises the following steps of: extracting various information of a communication radar map, and forming an acquisition failure ammeter communication data set by various data; based on the collection failure user set, forming a collection failure ammeter electricity utilization information data set on the collection monitoring page; extracting a characteristic value of the failed user data as a data characteristic of machine learning; the data set is divided, an artificial intelligence model is trained, and a model training effect and model tuning are tested; carrying out model training by adopting a machine algorithm random forest, and carrying out model evaluation on a model which is better trained; and the trained model is used to carry out batch marking processing on the actual acquisition failure ammeter fault reasons. Compared with the prior art, the failure electricity meter communication data is summarized and then serves as a training material to train the model, and the response speed of the trained model in practical application is remarkably improved compared with that in the prior art.