155 results about "Model testing" patented technology

The invention relates to safety compliance evaluation, in particular to a safety compliance evaluation system and method based on a multi-modal large model, and a functional module cluster, which are used for managing each functional module in the system, supporting cross-modal consistency detection and comparing answer consistency under different modals. The self-developed security system is responsible for execution of an assessment task, including resistance test enhancement and built-in multi-granularity data variation strategy, improves assessment robustness by simulating a real attack scene, realizes end-to-end automation from data generation, tested model test to risk assessment, supports configurable parameters and reproducible results, and has the advantages of high reliability and high reliability. The self-iteration and version management of the evaluation model are realized, and the self-iteration mechanism is deeply combined with a compliance label system, so that the risk identification capability under legal, ethical and industrial specifications can be continuously optimized; according to the technical scheme provided by the invention, the limitation that evaluation can be carried out only under a single mode and a single rule in the prior art can be effectively overcome.

The invention discloses a satellite orbit forecasting method based on deep learning physical constraint loss, and the method comprises the following steps: 1, carrying out the normalization preprocessing of input data, forming a training data set and a test data set, and constructing batch processing training data; and 2, performing dimension expansion on sample data points in each window in the batch processing data formed in the step 1, constructing a multi-dimensional feature space of the sample points, and forming a batch processing input data format capable of being introduced into the model. And 3, performing forward reasoning on the batch data formed in the step 2 by using a model, and obtaining a batch processing orbit prediction value output by the model at the next moment through a CNN lightweight spatial-temporal feature extraction module and a BiLSTM bidirectional time sequence neural network module. And 4, taking the track prediction value obtained in the step 3 and the truth value label in the training set obtained in the step 1 as input, calculating to obtain a loss value of a current training iteration batch through a multi-random learning loss module fusing physical constraints, and performing reverse updating of model parameters to complete model training. And step five, through the steps two to four, performing reasoning verification on the model by using the test set formed in the step one, and comparing with a truth value in the test set to obtain a model test result.

The invention provides a large model test case generation method, system and device based on multi-Agent collaboration. The method comprises the steps of obtaining product information of target product demand information; if the product demand information is incomplete, performing structured supplement on the extracted product demand information according to a slot filling technology to obtain complete product demand information; performing assembly according to the complete product demand information and the background information, and determining a first prompt word used for requesting the large model to generate a test case; and generating a first edition test case according to the first prompt word by using the plurality of agents, reviewing the first edition test case, generating an optimization suggestion, and fusing the first edition test case based on the optimization suggestion to generate a final test case. Through knowledge injection and structured requirements, the test case not only covers functional requirements, but also improves the accuracy of the test case; dynamic verification and correction of the use cases are achieved through cooperation of the multiple agents, the generation efficiency of the test use cases is improved, and the coverage rate is increased.

The invention discloses a model performance test method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence. The theoretical maximum lexical throughput of a target large language model is calculated based on the video memory bandwidth of a graphics processor, the model parameter quantity, the byte number corresponding to the quantization precision and the video memory bandwidth utilization rate; meanwhile, the benchmark performance throughput is obtained, a theoretical corresponding first concurrency number is calculated in combination with the theoretical maximum lexical unit throughput and the concurrency competition loss coefficient, then the model test is executed based on the first concurrency number to obtain the actual maximum lexical unit throughput and a corresponding second concurrency number, and a model performance test result is generated. The problems that in the prior art, due to the fact that manual testing is conducted depending on manual intervention, a continuous approaching attempt mode is adopted, a reasonable test starting point is not deduced in combination with hardware core bottlenecks and key parameters, evaluation is time-consuming and labor-consuming, the result is prone to being affected by artificial factors, and accuracy and consistency are poor can be solved.

The invention relates to the technical field of model testing, in particular to a model testing method and device and electronic equipment, and is used for solving the problems of long test data combination time consumption and low coverage rate of a model testing mode in the related technology. The multiple features are classified, a core feature set, a common feature set and an edge feature set are obtained, and the model can be a deep learning model; generating a test set of the model according to a differentiated data generation strategy of feature importance on the basis of the rule set of the service type adapted to the model, the core feature set, the common feature set and the edge feature set, and performing model performance test on the model by using the test set to obtain a test report; therefore, the efficiency of model algorithm testing is improved.

The invention relates to the technical field of risk management, and discloses a financial AI large model-oriented multi-dimensional test intelligent risk management and control method, which comprises the steps of integrating channel test, data test, AI test and AI large model test by constructing a financial AI large model test special database; traditional financial sensitive fields and AI test specific sensitive fields are accurately identified based on data consanguinity analysis, and dynamic desensitization is realized in combination with a lightweight desensitization engine; calling a preset risk scene library, defining a risk index system of each scene, dynamically calculating a risk level through a weighted scoring algorithm, and carrying out modeling by using a graph database to identify hidden linkage risks; triggering a grading response strategy according to the risk grade, and dynamically optimizing resource allocation; and after the test is finished, a multi-dimensional risk thermodynamic diagram and a root cause analysis report are generated, and processing experience is fed back to the scene library optimization strategy. According to the invention, the efficiency and safety of risk management and control can be improved.

The invention discloses a ceramic firing performance prediction method based on a GA-BP neural network, and the method comprises the steps: data collection and preprocessing, gray correlation analysis and screening of key process parameters, construction of a GA-BP neural network model, model training and verification, prediction model testing, and firing performance index prediction. By means of unique intelligent algorithm fusion, the problems of high experience dependence, high experiment cost, long period, low efficiency, time consumption, energy consumption, high efficiency and the like of a method for judging the ceramic firing performance through artificial experience in actual ceramic firing production are effectively solved. The technical problem that multi-target collaborative optimization is difficult to realize at the same time in the prior art is solved, remarkable advantages are shown in the aspects of prediction precision, intelligent degree and the like, the requirements of intelligent, green and high-quality ceramic firing process optimization are met, and the method has wide application prospects and important practical value.

The invention relates to the technical field of embedded artificial intelligence, and discloses an arc discharge detection model training method and system based on quantitative perception training, and the method comprises the steps: constructing an initial floating point precision neural network model for DC arc discharge fault detection, and defining a plurality of quantization functions for the model to convert floating point parameters; a FakeQuant pseudo quantization strategy is self-defined, a linear calculation layer in an original model is replaced with a quantifiable calculation layer, and a quantization operator is created through the quantifiable calculation layer; performing layer-by-layer differentiation quantitative perception training on the replaced model based on a training data set, and simulating a forward calculation behavior of a low-precision fixed-point type in actual reasoning while keeping floating-point precision back propagation through a quantization operator, so that the weight of the model is adaptive to a quantization error; and converting the trained model parameters into a target low-precision fixed-point type, and generating an end-side deployable arc discharge detection model through model testing and computing power evaluation, thereby realizing high-fidelity conversion from a floating-point model to a low-precision fixed-point model.

The invention discloses a privacy protection and robustness test method and system for large model fine tuning, and belongs to the technical field of machine learning security. The method comprises the steps that a three-layer distributed architecture comprising an edge server, a cloud server and a plurality of edge clients is constructed, the edge clients distill local privacy data and cooperate with the edge server to train a global model, and a candidate detection sample set is formed; screening a sample set based on the potential feature deviation evaluation index, and sending the sample set to a cloud server for vulnerability detection to obtain an optimal backdoor detection candidate sample set; and multi-trigger parallel and progressive trigger sequence backdoor implantation is respectively used for scenes of single fine tuning and multiple fine tuning of the large model, an optimal backdoor detection candidate sample set is combined with a preset trigger to generate a backdoor test sample set, the backdoor test sample set is mixed with a clean data set, and then the robustness of the backdoor test sample set is tested through fine tuning of the large model. Large model fine tuning and robustness testing of privacy protection can be realized in a heterogeneous model cooperative training environment.

The invention discloses a model training method and device, and the method comprises the steps: judging whether a target training model meets a model test condition or not according to a model test plan and the current training state of the target training model; if yes, copying the target training model to obtain at least one target test model; and performing model training on the target training model, performing asynchronous and parallel model testing on the target testing model, and judging whether the model training of the target training model is completed or not according to a model testing result. The resource utilization rate can be improved, and the model training efficiency is improved.

The invention relates to an air conditioner load prediction method based on a self-adaptive double-flow graph attention network, and belongs to the technical field of building energy conservation and intelligent control. The method comprises the following steps: collecting historical power and environmental data of an air conditioner, and after preprocessing and normalization, constructing an input sequence through a sliding window and dividing a data set according to time; a prediction model is constructed, and a causal graph learning module, a multi-scale graph structure learning module, a self-adaptive space-time attention module, an uncertainty quantization module and a self-adaptive sampling module are integrated; a training set and a joint loss function training model are adopted, and a load prediction result and uncertainty estimation are output through Monte Carlo Dropout during testing. According to the method, the dynamic causal relationship between variables and multi-scale space-time dependence can be adaptively learned, reliable uncertainty quantification is provided while the prediction precision is improved, and the method is suitable for intelligent regulation and control and energy efficiency optimization of the air conditioning system.

A robustness measurement method for LeNet-5 networks based on adversarial boundary constraints is presented, relating to the field of deep learning model testing. The main steps include: for each training sample, dynamically generating adversarial examples based on the model's current state during training iterations; constructing a composite loss function based on standard cross-entropy loss and dynamic boundary constraint loss; performing end-to-end training on all parameters of the LeNet-5 network; and using the overall approximate robustness boundary as a measure of model robustness after training. This method effectively improves the resistance of the LeNet-5 model to fast gradient sign-based adversarial attacks without altering the basic structure of the LeNet-5 network by designing a new loss function.

To provide a test precondition data derivation apparatus capable of easily deriving test precondition data corresponding to a test case.SOLUTION: Teacher data storage means stores teacher data used when generating, by machine learning, a model for deriving test precondition data that is either an SQL statement for registering data in a database of a system under test before execution of a test, or a file to be stored in a predetermined folder of the system under test before execution of the test, or both. Machine learning means generates a model for deriving the test precondition data by machine learning using the teacher data. Test precondition data derivation means derives test precondition data corresponding to a test case by applying the input test case to the model.SELECTED DRAWING: Figure 4

An application or software tool is used to test a device model with a physical device. The tool runs a test on the physical device via the network connection using the input of the test data. The test data includes the inputs and corresponding output values of the device model. The tool compares an output value of the physical device generated by the test with an output value of the device model to generate a result of the test. The test results are displayed on a display of the computing device.

This invention relates to the field of natural gas development and discloses a method, apparatus, storage medium, and processor for predicting the production of bottom-water gas reservoirs. The method includes: training a model based on a training set of production data from the bottom-water gas reservoir to obtain a first long short-term memory (LSTM) neural network model; optimizing hyperparameters based on a validation set to obtain a second LTM neural network model and a corresponding first error; testing the model based on a test set to obtain a corresponding second error; determining the generalization degree based on the first and second errors, and using the second LTM neural network model with a generalization degree meeting preset requirements as the target bottom-water gas reservoir production prediction model; and predicting the production of the bottom-water gas reservoir based on the target bottom-water gas reservoir production prediction model. This method captures long-term dependencies in time-series data using a LTM neural network, which can significantly improve the accuracy of bottom-water gas reservoir production prediction, thereby providing reliable support for oil and gas exploration and production decisions.

The invention discloses a closed data space model test self-adaption method based on double-stage feature whitening, and belongs to the field of model self-adaption, and the method comprises the steps: inputting each enhanced view in an enhanced data set composed of a first enhanced view and a second enhanced view into a data processing model, obtaining an intermediate layer activation feature and an output result respectively corresponding to each enhanced view; determining a first whitening feature corresponding to each intermediate layer activation feature, and determining a second whitening feature corresponding to each first whitening feature after projection mapping based on all the first whitening features; and performing parameter updating on a target data layer in the data processing model based on joint model loss constructed by the output result and the second whitening feature. According to the method, a two-stage feature whitening mechanism is introduced, so that the model can obtain better adaptive ability in a closed data space, and the generalization performance and reasoning stability of the model in a complex test environment are improved.

The invention relates to a model detection method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence. The method comprises the following steps: according to a first preset model, a to-be-processed data set and a first preset data set, updating an initial weight parameter of the first preset model to obtain a target weight parameter; then, according to the target weight parameter, a second preset data set and a first preset model, gradient information of the second preset data set is determined; and then, according to the gradient information, updating the to-be-processed data set to obtain a target data set. And further, training a second preset model according to the target data set to obtain a test model for attack test. Wherein the first preset data set does not comprise the abnormal training data, and the second preset data set comprises the abnormal training data. The method is used for testing the safety and robustness of the model.

The invention belongs to the technical field of intelligent medical big data disease risk prediction, and provides a VTE risk prediction method and system based on gradient increasing, and the method comprises the following steps: collecting a sample feature data group through a hospital system; preprocessing the collected sample feature data; performing correlation analysis on the processed sample feature data; calculating a risk threshold value of the numerical type sample features; establishing a machine learning model by adopting a gradient increasing algorithm; testing a model prediction effect; and performing VTE risk prediction by using a gradient increasing model. According to the method, feature data related to VTE diseases are doubly screened by combining feature importance ranking of the Pearson's correlation coefficient and the gradient increasing model, and linear and nonlinear correlation features are captured at the same time; the problems of prediction deviation and insufficient interpretability caused by one-sided feature screening and separation of threshold setting from diagnosis and treatment practice of an existing machine learning model are solved.

This invention provides a method for automatically generating test cases based on SysML state machine diagrams, comprising the following steps: S1, SysML state machine diagram parsing; S2, model conversion, converting the SysML state machine diagram into an intermediate model that is easy to analyze; S3, test path generation; S4, test data and test case generation; S5, test document generation. This invention also provides a readable storage medium. The beneficial effects of this invention are: it provides a method for automatically generating test cases based on SysML state machine diagrams, which can discover potential problems as early as possible in the development process, improve testing efficiency, reduce testing time and cost, and better meet the sufficiency requirements of testing.

The utility model discloses a test device for simulating the influence of aircraft taxiing and landing on foundation settlement, which relates to the technical field of foundation treatment and model test and comprises a trolley, a telescopic smooth guide rail, a thrust device, a vertical telescopic smooth platform, a buffer device, a monitoring sensor, a model runway and a model soil layer. The device creatively integrates a runway model box to reproduce a runway surface structure layer and a pile-net composite foundation, a taxiing landing device is adopted to simulate the taxiing landing process of an aircraft, and braking is achieved through a buffer device. A drainage device is embedded in the device, the sliding trolley is arranged in a front three-point mode to simulate a main landing gear of an airplane, and a 3-degree gliding angle is kept to be close to the real sliding condition. The test device not only considers the transverse impact force of an aircraft, but also performs adaptive design for different tire configurations, and provides a valuable simulation and monitoring means for exploring the dynamic characteristics of aircraft taxiing-landing integration on the pile-net composite foundation.

The invention provides a model testing method and device, electronic equipment, a storage medium and a program product, relates to the technical field of communication, and is used for improving the accuracy of model testing. The method comprises the steps of obtaining an output result of a to-be-tested model in a test communication environment and a confidence coefficient corresponding to the output result; wherein the to-be-tested model is a model used for executing a communication task, and the output result comprises communication control parameters output by the to-be-tested model; determining a communication performance index of communication based on the output result in the test communication environment; and according to the confidence coefficient and the communication performance index, determining whether a compatibility conflict exists between the to-be-tested model and the test communication environment.

This invention belongs to the field of image processing technology, specifically a spatial-context-based multi-branch image classification method. The invention includes data preprocessing, dataset partitioning, model training, model testing, and model performance verification. This invention employs a synergistic effect of spatial paths with rich spatial information and contextual paths with rapid downsampling, achieving a proper balance between speed and classification performance, while simultaneously realizing high accuracy and real-time performance.

The invention provides a privacy protection calculation method based on group learning and cardiovascular disease prediction, and the method comprises the steps: obtaining local medical data, carrying out the preprocessing, and dividing the data into a training set and a test set; training at least one base learner according to the local data characteristics and outputting a prediction probability; a Gaussian difference privacy mechanism is adopted to disturb prediction output; dynamically adjusting the privacy budget epsilon in an exponential decay mode according to the number of rounds in the training process; receiving prediction results after disturbance of all the clients and aggregating the prediction results by using a logistic regression model as a meta-learner; training a lightweight student model through knowledge distillation; the prediction capability is evaluated according to multiple performance indexes, and the influence degree of the differential privacy mechanism on the model performance is verified through a model test result under multiple epsilon values; and constructing an attack model to evaluate the protection effect of the privacy mechanism. On the premise of ensuring data privacy security, the cardiovascular disease prediction capability is improved, and the optimal balance between privacy protection and model performance is realized.

The invention discloses a large model test set evaluation method based on a project reaction theory, which belongs to the technical field of large language models, and comprises the following steps: (1) data preparation; (2) embedding and obtaining questions; (3) reference instance selection; (4) obtaining a model test result; (5) constructing a network structure; (6) model training; (7) Fisher information calculation is carried out; and (8) performing quality evaluation and parameter analysis on the test set. According to the method, the Fisher information, the difficulty parameter b, the distinction degree parameter a, the guess parameter c, the feasibility parameter d and other multi-dimensional indexes of each question are obtained, and the quality of each benchmark test is analyzed and evaluated based on the indexes, so that the quality of the benchmark test is analyzed more accurately and more comprehensively, and then the capacity of a large model and the quality of a test set are evaluated more accurately.

The invention belongs to the technical field of mechanical equipment fault diagnosis, and particularly relates to an equipment fault diagnosis method based on BO and an intelligent model, and the method comprises the following steps: S1, analyzing application scene data; s2, intelligent model selection of anomaly detection and fault identification; s3, optimization parameters are determined; s4, data set division; s5, determining model parameters based on Bayesian optimization; s6, testing an optimal parameter model; s7, determining an output fusion scheme of the anomaly detection and fault recognition model, and performing overall testing; s8, optimal parameter model application; s9, an engineer performs maintenance according to conditions; and S10, updating the model when the data type or scale changes. Test verification shows that the scheme has a good effect on diagnosis of various faults under variable working conditions.

The invention discloses a test method and device for an SQL statement generation model, equipment and a storage medium, and belongs to the technical field of computers. And in response to a model test request for the SQL statement generation model, converting the plurality of target test statements into a plurality of test SQL statements corresponding to the target test statements by using the SQL statement generation model. And executing the test SQL statement and the annotation SQL statement of each target test statement to obtain a first execution result of the test SQL statement of each target test statement and a second execution result of the annotation SQL statement, thereby increasing the test on the execution level of the SQL statement. The model test result is determined by combining the structural difference information between the test SQL statement and the annotation SQL statement of each target test statement and the execution result difference information between the first execution result and the second execution result, so that the structural difference of the SQL statements is considered, and the difference of the SQL statements in the execution result is also considered; and the accuracy of the obtained model test result is relatively high.

The application provides a model test system, method and related device of a multi-chip computing device. The system comprises an input module configured to obtain command line parameters and configuration parameters; a main control module configured to parse the command line parameters and the configuration parameters, and send an inference command to a performance stress test module; an API service starting module configured to start an inference service based on the parsed command line parameters; the performance stress test module is configured to construct a plurality of parameter combinations based on the parsed configuration parameters, perform inference testing based on each parameter combination respectively, and send corresponding inference results to the main control module; and an output module configured to output the inference results corresponding to each parameter combination. In the application, when testing, a user only needs to input command line parameters and configuration parameters, the main control module can schedule the API service starting module to configure environment variables, load a model to be detected, and send an inference request to an API inference service process based on each parameter combination respectively and start inference.

The present application relates to the technical field of composite structure design, and particularly relates to a composite pipe bending stiffness prediction method based on deep learning, a system thereof and a terminal, the composite pipe bending stiffness prediction method comprising: testing the bending performance of a layer-laid composite pipe under cantilever beam conditions by using a finite element model to build a basic data set; processing the basic data set to obtain a standard data set; constructing a neural network model, training the neural network model by using the standard data set, and obtaining a pipe bending stiffness prediction model; and outputting the bending stiffness by using the model. By simulating the composite pipe simulation to obtain the data thereof, the preprocessed data is divided into layer-laid data and pipe geometry data as independent variables, and the bending stiffness as a dependent variable, a sample data set is made, a pipe bending stiffness prediction model is constructed, the pipe bending stiffness can be calculated by using the basic data, and the low-cost and fast and accurate design of the composite pipe is realized.

The invention provides a large model testing system, method and device and a medium, and relates to the technical field of artificial intelligence, in particular to the technical fields of machine learning, deep learning, large models and the like. The large model to be tested is configured to execute a plurality of rounds of code iterative optimization tasks. The system comprises an execution unit configured to execute a code sample generated by a large model in a current round in a sandbox environment to obtain an execution result; the index unit is configured to calculate a performance improvement index based on the execution result, and the performance improvement index represents code generation performance changes of the large model among a plurality of rounds; the feedback unit is configured to select a target level from a plurality of preset diagnosis levels based on the performance improvement index, and the plurality of diagnosis levels respectively represent different feedback information detail degrees; and feedback information is generated based on the target level and the execution result, and the large model executes the next round of code iteration optimization task based on the feedback information.

The application discloses a KPI anomaly detection method fusing Bi-TCN and an adversarial VAE. The method comprises the following steps: S1, data preprocessing; S2, model initialization, random initialization and assignment of model parameters, and parameter setting; S3, model training; and S4, model testing, same data processing is performed on the test, a reconstructed error is obtained by inputting the trained model, and a threshold value generated by using an automatic threshold method is used to determine whether an anomaly exists. The application uses a variational autoencoder for adversarial training, the VAE is more robust to noise and abnormal values, and adversarial training can more efficiently amplify the reconstructed error of an input containing an anomaly to distinguish normal KPI data from abnormal KPI data.