767 results about "Reference modeling" patented technology

The invention relates to the field of chip verification, and provides a chip verification method, device and equipment, a medium and a chip. The method comprises the following steps: establishing a dynamic simulation verification environment of a to-be-verified design, and generating a plurality of test scenes and corresponding test excitation according to a design specification definition; sending the test excitation to a driver through a sequencer, converting the test excitation into a signal conforming to a to-be-verified design interface protocol by the driver, and transmitting the signal to the to-be-verified design; the monitor collects an input signal and an output signal of a to-be-verified design in real time, converts the input signal and the output signal into transactions, and transmits the transactions to the scoreboard; comparing the reference model with an output signal of the to-be-verified design to generate a dynamic simulation verification result; according to the dynamic simulation result, the state space is reduced, state space traversal is carried out on a design part which is not covered by dynamic simulation in the design to be verified, and a formal verification analysis result is generated; and evaluating the to-be-verified design according to the formal verification analysis result.

The invention provides a sparse feature constrained direct preference optimization method and device. The method comprises the following steps: inputting each preference sample pair into a reference model of freezing parameters to obtain a corresponding reference dynamic marginal value and a reference sparse activation feature; copying parameters of the reference model to obtain a dynamic strategy model, and inputting each preference sample pair into the dynamic strategy model to obtain a corresponding strategy sparse activation feature; constructing a mean square error as a sparse feature constraint according to the difference between the strategy sparse activation feature and the reference sparse activation feature, and taking a reference dynamic marginal value as a marginal value constraint; constructing a loss function based on the sparse feature constraint, the marginal value constraint and the reward function, performing parameter updating on the dynamic strategy model based on the loss function to obtain a preference model, and optimizing the preference of the model by introducing the sparse feature constraint and the marginal value constraint on an original DPO frame. The effect that the efficiency and the controllability are balanced at the same time is achieved.

There are provided a method for designing a spectacle lens and a related technique thereof, the method including: a modeling step of dividing a pattern into multiple models with a deviation from a normally worn state of the spectacle lens as decentering, in which a reference model, a decenter model, and a tilt model are prepared as the multiple model; a sensitivity calculation step of calculating a decenter sensitivity and a tilt sensitivity; and a design step of using a base curve value in the vicinity of a balance solution as a base curve of the spectacle lens, wherein a base curve value c [unit: diopter (D)] which is a curvature of a surface in a region H where the on-retinal non-convergence region is not provided on an object-side surface is taken as x-axis, and the decenter sensitivity and the tilt sensitivity [unit: diopter (D)] are taken as y-axis, and an intersection of a plot of the decenter sensitivity and a plot of the tilt sensitivity is taken as the balance solution.

Systems and methods for determining neural network brittleness are disclosed. For example, the system may include one or more memory units storing instructions and one or more processors configured to execute the instructions to perform operations. The operations may include receiving a modeling request comprising a preliminary model and a dataset. The operations may include determining a preliminary brittleness score of the preliminary model. The operations may include identifying a reference model and determining a reference brittleness score of the reference model. The operations may include comparing the preliminary brittleness score to the reference brittleness score and generating a preferred model based on the comparison. The operations may include providing the preferred model.

The invention discloses a nondestructive testing method and system and a data processing method, and relates to the technical field of nondestructive testing. Comprising the steps of obtaining initial detection data of a detected component; establishing at least one reference model according to the initial detection data, and obtaining simulation data corresponding to each reference model; determining target data according to the initial detection data and simulation data; and obtaining subsequent detection data, and comparing the subsequent detection data with the target data to form an evaluation result. According to the invention, high and low frequency mixed excitation signals are adopted, so that the penetrating power and resolution of detection are improved, surface defects and buried defects can be detected at the same time, and the comprehensiveness and accuracy of detection are remarkably enhanced; a picked complex response signal is decoupled by using an orthogonal phase locking method, a magnetic field signal and an electric field signal are effectively separated, the signal-to-noise ratio and the detection precision are improved, and accurate discrimination of defect types, depths and orientations is realized by establishing a reference model and acquiring analog data.

The invention discloses an optimization method and device for improving a near-end strategy based on a language model and electronic equipment, and the method comprises the steps: determining a plurality of answers corresponding to question data, and determining a first probability value of a single vocabulary in each answer based on the plurality of answers; determining a second probability value of a single vocabulary in each answer based on a reference model in the language model, and determining a KL divergence based on the first probability value and the second probability value; determining an instant reward for each answer based on a reward model in the language model and the target reward signal; and determining a target loss function based on the question data, the answer, the first probability value, the immediate reward and the KL divergence to update a policy model of the language model based on the target loss function. According to the method provided by the invention, the memory occupation problem in the near-end strategy training process can be relieved, and the memory usage is reduced, so that the concurrency is improved, and the answer generation and training process is accelerated.

The invention discloses an intelligent event studying and judging method based on machine learning, which comprises the following steps of: extracting features from multi-source event data, constructing an event fragment set with a uniform structure according to a preset time window, identifying fragments with stable features in the event fragment set as normal samples, constructing a reference model through an MCD algorithm, and analyzing the normal samples. The method comprises the following steps: extracting feature distribution in a normal state, comparing a to-be-analyzed event with a reference model, calculating feature offset, forming a behavior trajectory, determining an anomaly judgment range by adopting an elliptical envelope algorithm, and finally evaluating an event risk level and dynamically adjusting reference sample composition according to a continuous anomaly condition to realize intelligent identification and risk judgment of the event. According to the invention, dynamic perception and intelligent identification of complex events can be realized.

The invention discloses a cue word compression method and system based on TF-IDF and cross entropy, belongs to the technical field of large model cue word compression, and aims to solve the problems that redundant information is introduced into long cue words, the model efficiency is reduced and the cost is increased. To-be-compressed content is divided into sentences at the sentence level and then converted into embedded vectors, and the Euclidean distance is calculated in combination with problem vectors so as to screen related sentences; calculating a TF-IDF value at the word level through a word frequency and an inverse document frequency to extract keywords and recombine sentences; and selecting a reference model and a basic model at the Token level, identifying the key Token based on a cross entropy loss difference value, and splicing the key Token in sequence to generate a compressed cue word. According to the method, a complex calculation structure is avoided, the inference efficiency is improved while the semantic integrity is maintained, and the resource consumption is reduced.

The invention provides a distributed steer-by-wire vehicle stability control method. The method comprises the steps that a distributed steer-by-wire vehicle dynamics model, a two-degree-of-freedom reference model and a four-wheel steering angle distribution model are established; in the design of the yaw stability controller, a hybrid sensitivity # imgabs0 # robust control method is adopted, the uncertainty of the system is fully considered, and the robustness of the system can be considered while the target tracking performance is ensured.

The invention relates to a motor energy-saving control method for adaptive parameter adjustment. The method comprises the following steps: constructing an operation parameter sequence and comparing the operation parameter sequence with a preset reference model by collecting the rotating speed, the torque and the current signal of a motor in real time to generate a parameter deviation vector; based on a deviation threshold comparison result, carrying out noise processing on the overrun deviation to obtain a filtered sequence, and calculating a time delay degree in combination with historical parameters; calculating a compensation gain according to the time delay, fusing the compensation gain with the filtered deviation to generate a control instruction sequence, correcting the compensation gain through fuzzy control to obtain a stabilization parameter, and further generating a correction control instruction sequence; and finally, acquiring load data, fusing the load data with the correction instruction to obtain a system operation characteristic value, and generating an optimization state identifier when a judgment standard is met. By adopting the method, adaptive adjustment of the control parameters can be realized, the dynamic working condition is accurately responded, and the energy-saving efficiency of the motor under the variable-load working condition is improved.

The invention relates to a text-to-SQL (Structured Query Language) generation method and system based on large language model fine tuning. The method comprises the following steps: collecting a triple data set; generating a reasoning path for the triple data set, and obtaining an optimized data set comprising the triple data set and the corresponding reasoning path; selecting a basic model, and performing supervised fine tuning training on the basic model through the optimized data set to obtain a reference model; executing feedback through a database corresponding to an output result of the reference model, and performing direct preference optimization training on the reference model to obtain a generative model; and performing SQL generation on a to-be-generated text through the generation model to obtain a generation result. The method utilizes the advantages of the model language reasoning ability and the preference learning algorithm to improve the accuracy of the model in the aspect of text-to-SQL (Structured Query Language) generation tasks.

The invention discloses a 3D art model intelligent generation system and method based on multi-source data fusion, and relates to the technical field of computer aided design and graph generation, and the system comprises a multi-source data input module which is used for receiving a conceptual design drawing, text description and a reference 3D model; the feature extraction module comprises an image feature extraction unit used for extracting a style feature vector and a structure contour vector of the conceptual design drawing; and the text semantic analysis unit is used for outputting a key attribute tag vector described by the text. According to the 3D art model intelligent generation system and method provided by the invention, through multi-source data input including the concept design drawing, the text description and the reference 3D model, the design intention is comprehensively fused, the limitation of a single data source generation model is solved, and the dynamic weight distribution strategy and the closed-loop optimization mechanism of the 3D art model intelligent generation system and method are high in practicability. The geometric shape and style characteristics of the generated model are highly consistent with the design intention, and the workload of manual later correction is reduced.

The invention provides a preference alignment optimization method based on reward-driven selective punishment, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining intelligent question and answer training data, and constructing an intelligent question and answer training sample set which comprises a plurality of intelligent question and answer training samples; taking a to-be-optimized large language model as a strategy model and a reference model; and training the strategy model by using the intelligent question and answer training sample set, and measuring the offset amplitude of the strategy model before and after training by using the reference model to obtain an optimized large language model. According to the method, implicit reward signals in the model are introduced, preference data are divided into multiple categories according to implicit reward distribution generated by the model, a dynamic weight function is designed, differential weighted optimization is carried out on different categories of samples, reinforcement learning of high-quality samples and suppression of low-quality samples are achieved, and the method has the advantages of being high in robustness and high in robustness. Weight of low-quality or conflict samples is reduced while high-quality sample learning is enhanced, noise interference is suppressed, and overfitting is prevented.

A method for designing a spectacle lens and a related technique thereof, including: a modeling step of dividing a pattern into multiple models with a deviation from a normally worn state of the spectacle lens as decentering, wherein a reference model, decenter model, and tilt model are prepared as the multiple model; a sensitivity calculation step of calculating a decenter and a tilt sensitivity; and a design step of using a base curve value in the vicinity of a balance solution as a base curve of the spectacle lens, wherein a base curve value which is a curvature of a surface in a region where the on-retinal non-convergence region is not provided on an object-side surface is taken as x-axis, and the decenter and tilt sensitivities are taken as y-axis, and an intersection of a plot of the decenter sensitivity and the tilt sensitivity is taken as the balance solution.

The invention provides a chip verification method and device, equipment, a medium and a program product, and the method comprises the steps: receiving a first operation instruction of a to-be-verified module and a second operation instruction of a reference model in parallel, the reference model being a function model used for simulating the expected behavior of the to-be-verified module; searching a first instruction record matched with the first operation instruction from a request queue of the reference model, and searching a second instruction record matched with the second operation instruction from a request queue of the to-be-verified module; and determining a verification result of the to-be-verified module according to a comparison result of the first operation instruction and the first instruction record and / or a comparison result of the second operation instruction and the second instruction record. According to the invention, a structured and reusable universal chip verification method is realized, so that a tedious customized scoreboard design is replaced, and the efficiency and accuracy of verification work are greatly improved.

The invention discloses a construction product style migration method based on a diffusion model and structure control, and belongs to the field of computer-aided industrial design. Obtaining a structure reference STEP model and a target style image, and respectively extracting feature token sequences of the structure and the style; splicing the structure and style feature token sequence as a mixed condition vector of the conditional diffusion model; when the model diffuses in a submerged space, decoupled double-path attention layers are used for calculating cross attention output of structure and style features respectively, time-varying weights are added to serve as cross attention output of hybrid conditions for decoupling coding, and a 2D rendering graph conforming to a target structure and style is generated through gradual de-noising. The conditional diffusion model is run a plurality of times to generate a batch of candidate 2D renders. And screening out a qualified 2D rendering graph based on NURBS curved surface reconstruction. According to the invention, a structure-style dual-condition control mechanism is adopted to realize accurate balance between high-fidelity style migration and engineering constraint.

The present disclosure provides a model training method and device, and a data processing method and device. The method of the present disclosure uses a reference model to screen an existing instruction dataset for an instruction whose model fitting difficulty meets a preset condition, and the method can screen the instruction dataset for a challenging instruction having high model fitting difficulty as a seed instruction. A plurality of similar instruction samples are generated by means of expansion on the basis of the seed instruction, thereby obtaining more challenging instruction samples, a training set comprising the instruction samples and reference responses for the instruction samples is constructed, and knowledge distillation can be achieved by using the reference model on the basis of the existing instruction dataset, thereby obtaining a training set containing higher-quality instruction data. Furthermore, using the training set to train a deep learning model enhances the capability of the deep learning model to handle more complex and challenging tasks, and improves the performance of a trained target model.

The invention discloses a vehicle dynamics Kupman modeling and predictive stability expansion control method, and belongs to the technical field of vehicle dynamics modeling and safety technology control. The invention aims to provide a vehicle dynamics Kupman modeling and predictive stability expansion control method designed by combining a model predictive control method based on a closed-loop data promoted Kupman operator modeling method. According to the method, firstly, a linear two-degree-of-freedom reference model of the vehicle is designed, then a nonlinear closed-loop control system is established, Kupman lifting is carried out, a high-dimensional observation function is constructed, and finally, a linear MPC controller is designed. According to the method, linearization of vehicle dynamics is achieved through the Kupman model, the calculation complexity is remarkably reduced, the real-time performance of the controller is greatly improved, and a control system can rapidly respond under the dynamic working condition.

The invention relates to a verification method and device of a DMA controller, computer equipment, a readable storage medium and a program product. The method comprises the following steps: in response to a starting instruction of a verification script, extracting a test case of a to-be-verified DMA controller based on a pre-constructed verification environment, sending the test case to the to-be-verified DMA controller and a corresponding reference model according to a target path of the test case, and obtaining a first running result of running the test case by the to-be-verified DMA controller, and obtaining a second running result of running the test case by the reference model, comparing the first running result with the second running result, and generating a verification result of the DMA controller to be verified. The test case is automatically generated, so that the workload of manual writing is reduced, and the efficiency and quality of the verification process are remarkably improved.

The invention discloses a non-cooperative spacecraft attitude estimation method and system based on a laser radar, and the method comprises the steps: obtaining a scanning point, observed by the laser radar, of a non-cooperative spacecraft, carrying out the attitude initialization of the scanning point, obtaining a scanning point cloud, constructing a reference model of the non-cooperative spacecraft, and carrying out the attitude estimation of the non-cooperative spacecraft. The scanning point cloud is matched with the reference model to obtain a matching result, attitude tracking is performed on the matching result based on an iterative nearest point algorithm and a Kalman filter to complete attitude estimation of the non-cooperative spacecraft, and the problem of attitude tracking of the non-cooperative spacecraft of an airborne LiDAR system is solved. The limitation of an attitude initialization method based on template matching and point cloud registration is overcome, and the relative state of the non-cooperative spacecraft can be accurately estimated.

The invention discloses a power industry large model continuous pre-training method and system based on dynamic self-constraint, and belongs to the technical field of artificial intelligence, and the method comprises the steps: obtaining industry pre-training corpora and instruction training corpora, dynamically adjusting the mixing ratio of the industry pre-training corpora and the instruction training corpora through a curriculum-type strategy, and obtaining an industry pre-training corpora and an instruction training corpora; obtaining a dynamic mixed data set; configuring a reference model based on the dynamic mixed data set, and training a target power industry large model by adopting a differential loss function and the reference model for different types of data in the dynamic mixed data set; according to the differential loss function, self-adaptive KL divergence is calculated according to inter-partition optimization logic, and the self-adaptive KL divergence is adopted to construct a loss function; and obtaining the probability of the reference model through an online reasoning framework, and enabling the training to be continuously carried out based on the probability of the reference model. According to the method, the knowledge conflict problem in professional field training is effectively solved, and the generality of the model is kept while the professional property of the power field is improved.

The invention discloses a photovoltaic module construction quality identification method based on computer vision and three-dimensional modeling. The method comprises the following steps: S1, collecting an original image sequence of a photovoltaic construction site and carrying out image preprocessing; s2, outputting a semantic tag graph and a two-dimensional space boundary graph through an improved BiSeNet network; s3, using a mask constraint SIFT algorithm to extract cross-view key point features, generating a dense three-dimensional point cloud according to multi-view stereo reconstruction, and outputting a structure three-dimensional geometric model; s4, mapping the semantic tag graph and the two-dimensional space boundary to the structure three-dimensional geometric model to obtain a semantic three-dimensional model; s5, performing spatial registration on the semantic three-dimensional model and the reference model, and calculating spatial error parameters of the photovoltaic module; s6, performing compliance discrimination on the spatial error parameters, and outputting a quality discrimination result; and S7, generating a construction quality evaluation report. According to the invention, automatic identification and accurate evaluation of the photovoltaic construction quality are realized, and the detection efficiency and the discrimination accuracy are improved.

A method for monitoring the positioning of the teeth including production of a three-dimensional digital initial reference model of the arches of the patient and, for each tooth, definition, from the initial reference model, of a three-dimensional digital reference tooth model; acquisition of updated image of at least one two-dimensional image of the arches in actual acquisition conditions; analysis of each updated image and production, for each updated image, of an updated map; optionally, determination, for each updated image, of rough virtual acquisition conditions approximating the actual acquisition conditions; searching, for each updated image, for a final reference model corresponding to the positioning of the teeth during the acquisition of the updated image, for each tooth model, comparison of the positionings of the tooth model in the initial reference model and in the reference model obtained at the end of the preceding steps to determine the movement of the teeth.

The invention relates to a motor simulator current tracking control method and system, and the method comprises the steps: building a reference model of a motor simulator test system which comprises a tested motor drive module and a motor simulator connected with the tested motor drive module; performing simulation comparison on the branch current of each parallel branch and the model output reference current with single-phase total current feedback, setting a hysteresis coefficient of hysteresis comparison according to a hysteresis comparison principle, and generating a hysteresis comparison result; according to the branch states of the tested motor driving module and the motor simulator, current increase and decrease of a parallel branch on the inverter side of the motor simulator are controlled in combination with a hysteresis comparison result, commutation is carried out after current increase and decrease at a single time in each period, and multi-branch fixed-frequency phase-dislocation current control is achieved. By adopting a current hysteresis comparison control method with total current feedback compensation and a fixed-frequency phase-dislocation current control algorithm, the high-frequency and high-power working requirement of the motor simulator is met, and the current tracking precision and the dynamic response speed are improved.

Embodiments of the disclosure provide a solution for machine learning model evaluation. The solution includes: obtaining a target answer to a test question generated by a target machine learning (ML) model; obtaining a plurality of reference answers to the test question generated respectively by a plurality of reference ML models; determining respective professional levels of the plurality of reference ML models in answering the test question; and generating an evaluation result on correctness of the target ML model in question answering based on the target answer, the plurality of reference answers and the respective professional levels of the plurality of reference ML models.

The invention discloses a VPSC model construction method based on an ABAQUS sub-model, and belongs to the field of computational material science and engineering.The VPSC model construction method comprises the steps that a global model is established according to the size of a part, and an ABAQUS input file is generated; selecting a local area of a part in the global model, determining a unit number of the area, constructing a crystal plasticity constitutive model through UMAT, and adding a plurality of randomly oriented crystal grains to the local area in an ABAQUS input file to form a reference model; grain nodes and orientation information in the reference model are read through an ABAQUS script, a grain model is created, corresponding orientation is given to sub-models, sub-model boundary conditions consistent with global model deformation are applied to grain geometric boundaries, and a VPSC model is constructed. According to the method, the grain orientation is locally refined through the sub-model technology, and the VPSC sub-model is automatically generated in combination with the script, so that the calculation complexity is remarkably reduced.

A method, apparatus, and / or system for soft defects modeling of measurements using time-domain reflectometry. Electro-Optic Sampling based Time-Domain Reflectometry (EOS-TDR) may quickly detect soft defects in a chip under test. For example, EOS-TDR may detect soft defects in each pin from a trace-structure point at a relatively high resolution. To interpret the results in a time sensitive manner, a reference model for chips may be established from chips that are known to have met the expected quality standards. Through automated analysis of the features of the device under test waveform, soft defects of a chip may be detected that would be otherwise undetectable under time constraints, temperature variations, applied current variations, applied voltage variations, vibration variations, moisture variations, or any other kind of possible variation.

The invention discloses a deep neural network model freezing training optimization method based on tensor similarity. According to the method, a reference model generation module regularly generates a single-layer reference model to support real-time tensor similarity evaluation; the tensor similarity calculation module generates a normalized Gram matrix by using the activation output, and accurately evaluates the stability of the active layer; the freezing decision module can smooth instantaneous fluctuation in an evaluation process based on a moving average value of tensor similarity so as to make a steady freezing decision; the tensor I / O module caches the forward propagation result of the freezing layer, and repeated calculation is avoided. According to the method, the calculation amount of back propagation and forward propagation in the deep neural network training process can be effectively reduced, and the calculation resource utilization rate in the training process is improved. The method is remarkably superior to an existing freezing method in deep neural network model training tasks such as image classification, target detection and image segmentation, and the training stability and efficiency are improved while the final precision of the model is not sacrificed.

The invention provides a clock tree verification method and device, a computer readable storage medium and computing device.The clock tree verification method comprises the steps that a connection reference model is constructed, and the connection reference model comprises information of clocks in a clock tree, the connection relation between the clocks and the initial value of a control signal; the information of the clock comprises a reference period and a reference burr expected value of the clock, and each control signal is used for controlling a connection relation between at least one clock; randomly updating the value of the at least one control signal, and updating the reference period and the reference burr expected value according to the updated value of the at least one control signal; and detecting each clock according to the updated reference period and the reference burr expected value. The invention provides a scheme for improving clock tree verification convenience and accuracy.

The invention relates to the field of function verification of digital integrated circuits, and discloses an atomic-scale computing chip verification method based on a UVM. The method comprises the following steps: firstly, building a UVM simulation verification platform comprising each core component; then, a software program is used for generating test excitation with physical significance, wherein the test excitation comprises information such as configuration, functional, pseudo potential and track files; then respectively inputting the test excitation into the established UVM verification platform and the software reference model; the UVM verification platform applies the test excitation to the to-be-tested design and outputs a calculation result; the reference model also correspondingly generates a reference result; and finally, comparing the output of the to-be-tested design with the output of the reference model, and verifying the correctness of the function of the atomic calculation chip through visual auxiliary analysis and quantitative error analysis. Through excitation generation based on physical constraints and an effective error evaluation method, the function verification problem of the atomic-scale computing chip can be solved.