16 results about "Validation error" patented technology

The present invention relates to the field of large model application technology, and discloses a code error repair method and system based on a large model. The method includes: deploying several large models dedicated to coding and initializing model scoring; continuously monitoring the error log of the target program, obtaining the current error information, and pulling the error code file corresponding to the current error information in the code repository; constructing a code repair prompt word, sending it to the large model with the highest model score, and guiding it to repair the error code; deploying the updated code file in a test environment, and using a request matching the error information to request replay of the updated code to verify whether the error is successfully repaired; if so, updating the target program according to the updated code file, and positively adjusting the model score of the currently used large model; if not, excluding the large model that failed to be repaired. The present invention realizes the automation of the entire process from error monitoring, code acquisition, error repair to deployment verification, reducing manual intervention.

The present application relates to a cross-validation training method, apparatus, and equipment for a large-scale flow field proxy model. The method comprises: determining the range of design variables and performing normalization processing; performing a recursive permutation evolution experimental design on the design variables, generating sample points, and performing numerical simulation to obtain flow field data as the response values of each node in the flow field; constructing a proxy model for each node in the large-scale flow field, and calculating the corresponding basis function matrix of the proxy model based on the same shape parameters; calculating the sum of squared cross-validation errors for the entire flow field by constructing a second intermediate matrix related only to the shape parameters and design variables and a global matrix related only to the node response values, and optimizing the shape parameters with minimizing the sum of squared errors as the optimization goal, until convergence and output of an optimal proxy model based on the optimal shape parameters to perform the flow field prediction task. This method can achieve rapid and reliable verification of large-scale flow field proxy models and efficient and accurate flow field prediction.

Techniques are described for quickly rerouting traffic to an application hosted on a first Virtual Private Cloud (VPC) location. In the event of an error in routing traffic to the first VPC portion traffic can be rerouted to a second VPC portion. The first and second VPC portions can be different portions of the same VPC or could be different VPSs. The techniques include steps for calculating a predetermined route to the second private virtual cloud location. The techniques further include steps for monitoring data for detecting an error in the first cloud location. The steps further include detecting a potential error based on the monitored data, and also verifying that the potential error is in fact a valid error. In response to verifying that the error is a valid error, further steps include performing a fast reroute of traffic to the second cloud locations along the predetermined route.

The present application discloses a test data management method, apparatus, computer device, and storage medium, including: collecting the functional architecture of a target terminal, where the functional architecture is composed of multiple functional units; configuring verification rules for each functional unit based on the functional architecture and a preset rule database, where the rule database is used to store verification rules corresponding to various types of functional units; extracting test data of the target terminal, and splitting the test data according to each functional unit to generate multiple functional test data; respectively performing data verification on the corresponding functional test data according to the verification rules of each functional unit; and when any one of the functional test data fails to pass the verification, sending a verification error message to the target terminal. This avoids the problem of missing test scenarios caused by only focusing on test results, improves the coverage of test scenarios in test verification, and makes the test results more real and accurate.

The invention relates to the technical field of large model application, and discloses a code error repairing method and system based on a large model, and the method comprises the steps: laying a plurality of large models special for coding, and initializing model scores; continuously monitoring an error log of the target program, obtaining current error information, and pulling an error code file corresponding to the current error information in a code warehouse; constructing a code repair prompt word, sending the code repair prompt word to the large model with the highest model score, and guiding the code repair prompt word to repair the error code; deploying an update code file in the test environment, performing request playback on the updated code by using a request matched with the error information, and verifying whether the error is successfully repaired or not; if yes, updating the target program according to the update code file, and positively adjusting the model score of the currently used large model; and if not, excluding the large model which fails to be repaired. According to the method, the whole process automation from error monitoring, code obtaining, error repairing to deployment verification is realized, and manual intervention is reduced.

The application provides a large-span steel structure closed loop analysis method and system, the analysis method comprises the following steps: constructing a finite element model and a design parameter space, identifying key design parameters and key regions affecting the first N modal strain energy of the structure through initial sensitivity and modal analysis, and establishing a mapping relationship between the two; according to the parameter sensitivity contribution, a non-uniform sampling strategy is formulated, the sampling is encrypted in the high sensitivity parameter subspace, an initial sample set is generated, based on the sample and the finite element response, a hierarchical mixed kernel Kriging surrogate model containing a global long-range kernel and a multi-region local short-range kernel is constructed, the local kernel weight increases as the input point approaches the high sensitivity region, the iteration termination and the performance limit state are set, the composite point adding criterion integrating the prediction variance, the limit state distance and the cross-validation error is used to iteratively update the sample and the model, and after the condition is met, the structure performance analysis is completed by using the surrogate model.

The invention relates to the technical field of data processing, in particular to an accelerometer vibration rectification error test system and method, and the system comprises a self-adaptive signal collection module, a bidirectional cooperative transmission module, a composite modal statistics module and a cross validation error modeling module. The adaptive signal acquisition module generates a feature mark, and the feature mark drives the bidirectional cooperative transmission module to adjust the depth of a buffer area and injects a timestamp verification identifier; a composite modal statistics module selects a wavelet packet time-frequency fusion algorithm based on a timestamp missing proportion, and aligns a physical cycle segmentation window to extract a time-frequency entropy mean statistical magnitude; a cross validation error modeling module fuses statistics and vibration parameters to generate a quantitative model, and a confidence evaluation factor distribution diagram triggers double-path feedback. Invisible window optimization improves data quality in a key period, and frequency domain weight redistribution optimizes a feature extraction strategy. The problems of insufficient dynamic data real-time processing efficiency and transient feature extraction misalignment caused by data acquisition and analysis function separation in the prior art are solved.

Digit validation check control for instruction execution. A process receives an instruction to perform one or more operations using one or more input values. The instruction includes a "No Validation" indicator to check whether the digit validation check control is enabled for instruction execution. The process executes the instruction based on the "No Validation" indicator, including determining whether the digit validation check control is enabled for instruction execution and performing the processing based on this determination.Based on the "No Validation" indicator, which is set to a defined value, the "Digit Validation Check" control is activated, and the processing includes forcing a "Digit Validation Error" indicator, which is output by the execution to indicate "No Digit Validation Error" with respect to the at least one input value.

This disclosure relates generally to system and method to track performance of an AI application during development and production lifecycle. Libraries and tools that are used to develop an enterprise AI application are rapidly expanding across vendors and open-source community. These libraries are coming up with new features and breaking changes with new versions making it difficult to application developer and enterprise runtime executor. The method of the present disclosure receives an enterprise artificial intelligence (AI) application and corresponding software environment as input generate a logger plug-in using one or more fine-tuned large language models (LLMs) based on the plurality of instructions provided by the one or more instructors. Additionally, the logger plug-in is utilized to validate correctness of errors and then the logger plug-in is executed with the enterprise AI application to log and track performance of the enterprise AI application.

This invention provides a deep learning-based method for predicting the deformation of foundation pit support piles, belonging to the field of foundation pit engineering safety monitoring technology. The invention establishes a monitoring network to continuously collect multi-dimensional feature data and horizontal deformation values ​​of support piles, recording timestamps; sets the time window length and prediction step size, generates samples by sliding along the time axis, and divides them into training and test sets; constructs an AM-TKAN initial model, determines the type and initial range of hyperparameters to be optimized, and generates several hyperparameter combinations as individuals; trains the model corresponding to each individual on the training set and calculates the error, iteratively updates the hyperparameters according to the Hippo optimization algorithm, and outputs candidate optimal hyperparameter combinations that meet the error requirements; uses a validation set to select the model with the smallest validation error as the final prediction model; inputs real-time data into the final prediction model and outputs the deformation prediction results. This invention achieves high-precision prediction of the horizontal deformation of foundation pit support piles.

A method for processing sparse data sources, comprising: (a) retrieving data from the sparse data source to form a basic data set, wherein the sparse data source includes a plurality of patient records containing patient mortality data, and each of the patient records includes at least one unpopulated data field corresponding to a medical measurement; (b) dividing the basic data set into two parts: a first part including a training data set at a predefined ratio of the basic data set, and a second part including a validation data set at a predefined ratio of the basic data set; (c) analyzing the training data set to co-model variable relationships using a non-linear function approximation algorithm repeatedly applied to the records of the training data set for the purpose of obtaining a trained model and a measurement prediction protocol for populating the unpopulated data fields within the training data set; (d) calculating predicted values of measurement data for the unpopulated data fields using the measurement prediction protocol; (e) populating the predicted values into the records within the training data set; (f) analyzing the training data set based on a predefined disease state in the known patient records of the basic data set to form a phenotype model associating the patient's phenotype data with the probability of the disease state in the patient records of the training data set; (g) populating the predicted values into the records within the validation data set; (h) validating the validation data set using the phenotype model, and determining a validation error including the probability of correctly predicting the patient's phenotype associated with the probability of the disease state in the records of the validation data set; (i) repeating steps (c) through (h) to minimize the validation error, and calculating and predicting a high-probability disease state phenotype for each patient record in the basic data set. Records within the validation data set include the patient's phenotype data associated with the probability of the disease state.

This invention discloses an intelligent early warning method, device, and storage medium for offshore wind turbine foundation scour, belonging to the field of structural safety early warning technology. The method includes the following steps: obtaining the foundation structural parameters of the offshore wind turbine and establishing a finite element model; performing foundation scour simulation and collecting time-domain data of pile top displacement and pile nodal stress; establishing several surrogate models, dividing the time-domain data into training and validation sets, and obtaining training and validation results; using two validation indices, validation error and leave-one-out error, to select the best surrogate model; using different scour conditions as input parameters to obtain the wind turbine foundation response, comparing it with a set safety early warning threshold, and issuing a safety alarm if the wind turbine foundation response reaches or exceeds the safety early warning threshold. This invention can solve the problems of traditional experimental methods, such as the difficulty in reproducing initial conditions, the influence of experimental scale on measurement results, and high costs, as well as the shortcomings of finite element simulation in achieving full coverage of scour conditions.

The application relates to the technical field of data processing, in particular to an accelerometer vibration rectification error test system and method, which comprises an adaptive signal acquisition module, a bidirectional collaborative transmission module, a composite modal statistical module and a cross-validation error modeling module. The adaptive signal acquisition module generates a characteristic mark, the characteristic mark drives the bidirectional collaborative transmission module to adjust the buffer depth and inject a timestamp verification mark; the composite modal statistical module selects a wavelet packet time-frequency fusion algorithm based on a timestamp missing ratio, extracts time-frequency entropy mean statistics by aligning a physical cycle division window; the cross-validation error modeling module fuses the statistics and vibration parameters to generate a quantitative model, and a confidence evaluation factor distribution diagram triggers a double-path feedback: a stealth window optimizes and improves the data quality in a key period, and a frequency domain weight re-distribution optimizes a feature extraction strategy. The application solves the problems of insufficient dynamic data real-time processing efficiency and inaccurate transient feature extraction caused by the fragmentation of data acquisition and analysis functions in the prior art.

A parameter validation result generation method, comprising: receiving a request for calling a cloud application programming interface (API) by a client; on the basis of an MVC validation framework, validating the request and obtaining a validation result; when the validation result comprises a validation error message, finding a corresponding target error code enumeration on the basis of the validation error message and a preset error code enumeration list, each error code enumeration in the error code enumeration list comprising an error code and a validation error message corresponding to the error code; and, on the basis of the target error code enumeration, returning an error code of the request and the validation error message to the client. Thus, by means of binding a validation rule, the validation error message, and the corresponding error code, a corresponding error code and a validation error message are automatically generated for a validation failure result of each field. Further disclosed are a parameter validation result generation apparatus, a validation platform, and a storage medium.

A system for effort estimation in agile software development using multi-stage evolutionary stacking, consisting of: a data acquisition module configured to retrieve software effort records from one or more data set repositories containing historical data from software development projects with characteristics and actual effort values; a data preprocessing module that is operationally connected to the data acquisition module and is configured to receive the aforementioned software effort data sets from the data acquisition module, cleans the received data by removing inconsistencies with missing target values, and normalizes numerical input characteristics to a common range; a first-level ensemble module connected to the data preprocessing module, wherein the first-level ensemble module comprises a variety of heterogeneous basic learners, including a Random Forest model, a Support Vector Regression model, and an Extreme Gradient Boosting model, which generate predictions from each of the heterogeneous basic learners using the preprocessed data sets received from the data preprocessing module; A genetic algorithm optimization module connected to the first layer's ensemble module, configured to: encode weights as a normalized real-valued vector for each of the heterogeneous base learners; apply a fitness function to minimize the mean squared validation error and derive an optimal weight vector; assign optimized weights to the predictions of each of the heterogeneous base learners; and generate weighted predictions based on the optimized weights. a second-level meta-learning module connected to the optimization module of the genetic algorithm, configured to receive the weighted predictions from the optimization module of the genetic algorithm, processes the weighted predictions using a deep multilayer perceptron neural network to learn complex patterns and nonlinear interactions, and generates a final effort estimate for the software; an output processing module connected to the second-level meta-learning module, configured to receive the final effort estimate for the software and process and visualize the data to improve user understanding; and a user interface connected to the output processing module to receive the processed final effort estimate for the software, wherein the user interface is configured to display the processed and visualized final effort estimate for the software.

The present invention relates to the field of computer technology, and provides an ARP blocking verification method, system, device and storage medium. The method includes the following steps: S1: Block at least one host device of each different type in the local area network to be verified in advance; S2: Capture the mirror traffic data packets formed by the mirror traffic of all host devices in the local area network to be verified; S3: Analyze the mirror traffic data packets to verify whether the blocking situation conforms to the expectation: if so, the blocking verification is correct, otherwise, the blocking verification is incorrect. In a complex network environment, it can better verify the ARP blocking situation in the entire local area network, without the need for manual verification of the ARP blocking situation in the entire local area network, improving the accuracy of the ARP blocking verification result, and can better know the quality of the ARP blocking effect.