295 results about "Value set" patented technology

The invention discloses a MoE large model reasoning optimization method and device based on expert module dynamic scheduling, and relates to the technical field of large model reasoning optimization. The method comprises the steps of performing pre-reasoning through a MoE large model according to a dialogue text data set to obtain a first evaluation score set; performing value optimization on the GPU candidate expert quantity value set through an expert cache hit rate calculation method to obtain a GPU candidate expert optimized value set; cPU expert module dynamic scheduling is carried out based on the relation curve of the expert score ascending value-the transmission expert number, and a CPU expert scheduling index list is obtained; gPU expert module dynamic supplementation is carried out according to the first evaluation score set and the CPU expert scheduling index list, a GPU expert supplementation index list is obtained, re-reasoning is carried out through a MoE large model, and a reasoning result text data set is obtained. The invention relates to a large model reasoning optimization method for balancing model reasoning speed and output result quality of a MoE large model.

The invention relates to a data query method and device based on a natural language and a medium, and belongs to the technical field of data analysis. Receiving a natural language query input by a user; constructing task planning guide information based on the natural language query, and inputting the task planning guide information into the large language model to identify a target task type corresponding to the natural language query; when the target task type is a data query type task, determining a query target index corresponding to the natural language query from an index library; on the basis of a DSL dimension normal form corresponding to the query target index, dimension fields related to semantics of the natural language query and matched enumeration values are recognized from the multiple dimension fields and enumeration value sets; and constructing a structured intermediate representation, generating an executable code corresponding to the natural language query, and executing the executable code to obtain a corresponding structured query result from a data source. According to the method and the device, the accuracy and the interpretability are considered, and meanwhile, the SQL generation flexibility is improved.

The invention discloses a cold storage temperature optimization control system and method for a chilled water storage air conditioning system, and relates to the technical field of chilled water storage air conditioning management.The method comprises the steps that 1, a plurality of operation parameters in the system operation process are collected; performing time synchronization and abnormal value elimination on the operation parameters to form a dynamic operation data set; and 2, judging the current hydraulic circulation state based on the dynamic operation data set, and outputting a circulation deviation signal when any index of the total flow of the system, the flow distribution of each branch or the temperature difference change rate of the supplied water and the returned water exceeds a set corresponding threshold value. By constructing the dynamic operation data set and monitoring key parameters such as the system temperature, the flow and the water pump operation state in real time, a unified multi-dimensional data basis is established, support is provided for follow-up operation state recognition and control strategy execution, and the real-time performance and accuracy of control are remarkably improved.

The invention discloses a cross-device memory dynamic complementary utilization method based on CXL protocol interconnection and a computer system. The method comprises the following steps that an ECC module of a memory detects data in the data transmission process; when a data error is detected, the error is positioned by the ECC module, the error counting module corresponding to the position is updated, and an error report is generated; when the counting value of the error counting module reaches a set threshold value, the area corresponding to the position is marked as a failure area; evaluating the states of all memories based on the spatial dimension, the health dimension, the time delay dimension and the load dimension, and selecting a target memory according to an evaluation result of the states of the memories; distributing a new logic address for the failure area based on the error report and a selected target memory; the logic address is transmitted to a target memory through a CXL protocol, and an address mapping table is updated at the same time; and after the mapping is successful, migrating the data of the failure area to the healthy area of the target memory by the system through a CXL protocol.

The invention relates to an automatic interface test management method and device, equipment and a medium, and the method comprises the steps: obtaining an interface document of a to-be-tested interface, carrying out the analysis of the interface document, generating interface case data containing the primary configuration content of an interface case, and transmitting the interface case data to a server; performing execution grouping on the interface cases configured with the input parameter value set, generating a case execution group based on interface grouping information and a preset test execution strategy, setting an execution sequence and an exception handling rule for the case execution group, and generating a corresponding execution link at the same time; calling an execution link based on the trigger request to sequentially use each group of request content execution cases in the input parameter value set to execute the interface cases in the group, and generating corresponding case execution results; high automation and intelligentization are achieved in the whole interface test process, manual intervention is remarkably reduced, the test coverage rate is increased, the problem response speed is increased, and therefore the working efficiency of testers is effectively improved, and the human error risk is reduced.

The present invention discloses a performance evaluation method and device for an automatic and rapid deployment and retraction tent. The method comprises: collecting a performance index test data set for the automatic and rapid deployment and retraction tent; the performance index test data set comprises an deployment and retraction index test data set, an efficiency index test data set, and an environmental adaptability index test data set; each type of index test data set comprises several types of index test data sequences; preprocessing the performance index test data set to obtain a preprocessed data set; performing classification evaluation processing on each type of index test data set in the preprocessed data set to obtain a sub-index test value set; and performing fusion evaluation processing on the sub-index test value set to obtain a performance evaluation result value for the automatic and rapid deployment and retraction tent.

The invention discloses a micro-grid safety monitoring method and system based on attack detection data, and relates to the field of power grid safety monitoring, and the method comprises the steps: outputting a state vector of each node; obtaining a predicted attacked result of each node; outputting an estimated value set and a measured value set of the operation data of each prediction node; and obtaining an attack detection result of each prediction node and executing corresponding operation. According to the method, the data channel safety degree and the multi-dimensional evaluation parameters are fused, a state estimation and attack probability recognition mechanism combining extended Kalman filtering and a time sequence model is introduced, a hierarchical response mechanism is set according to attack levels, the elastic defense and active recovery capability of the system under different attack intensities is improved, and stable operation of the micro-grid is guaranteed.

The embodiment of the invention discloses a front-end page code generation method and device, electronic equipment and a medium. A specific embodiment of the method comprises the steps of performing multi-source data fusion on a multi-source front-end page design draft set and user historical page browsing behavior information to obtain front-end page fusion design information; performing element visual semantic extraction on the multi-source front-end page design draft set to obtain a page text element information set and a page design visual element information set; constructing a page behavior semantic knowledge graph; generating an initial front-end page code; code interaction adjustment is carried out after code adjustment is carried out on the initial front-end page code, and a front-end interaction code is obtained; performing multi-dimensional page evaluation on the front-end page to obtain a page multi-dimensional evaluation value set; and performing code dynamic adjustment and optimization on the front-end interaction code. According to the embodiment, the front-end page code can be automatically and accurately generated, the code generation efficiency and adaptability are improved, the page error with a design draft is reduced, and the user interaction experience feeling is improved.

A logical expression acquisition unit (502) acquires a logical expression that represents a condition under which searchable encryption is permitted, by a logical structure with a plurality of attribute values and a logical operator. An encrypted tag generation unit (504) sets a secret sharing value which reflects the logical structure, to each of the plurality of attribute values, and generates an encrypted tag to be used for the searchable encryption, using a plurality of secret sharing values set to the plurality of attribute values.

The invention discloses an interactive intelligent customer service robot interaction method and system based on a large language model, and relates to the technical field of artificial intelligence. A pre-trained BERT model is adjusted according to preprocessed text data to obtain an intention entity recognition model; predicting dialogue intention category probability distribution and a dialogue entity label sequence of the interaction object by utilizing an intention entity recognition model, constructing a causal relationship graph among dialogue intention categories, dialogue entity labels and behaviors, recognizing the dialogue intention category probability distribution and the dialogue entity label sequence of the interaction object in real time, and setting the dialogue intention category probability distribution and the dialogue entity label sequence as target demand variables; determining a value set of the target demand variable, calculating probability distribution of the value set according to the causal relationship graph, taking a value of which the probability is greater than a probability threshold as a main demand of an interaction object, retrieving a solution related to the main demand to generate a dialogue reply, and obtaining feedback of the interaction object to dynamically adjust a dialogue strategy. And intelligent customer service interaction work is realized.

The invention relates to the technical field of pool accumulated water processing, and discloses a pool accumulated water monitoring method and system based on an overflow gap, and the method comprises the steps: comparing a liquid level value digital signal with a threshold value set by a historical liquid level value digital signal, and outputting an overflow monitoring signal if the liquid level value digital signal is higher than the threshold value; receiving an overflow monitoring signal, collecting real-time data of an overflow port pool, constructing an overflow port simulation model based on a historical overflow port pool signal, and inputting the real-time data into the overflow port simulation model for simulation; judging the state of the overflow port pool according to the simulation result, formulating an adjustment scheme according to the state of the overflow port pool, inputting the adjustment scheme into the overflow port simulation model for simulation, and performing iteration in sequence until the data of the overflow port pool is qualified. The system comprises an overflow signal acquisition module, an overflow simulation module and a scheme making module. Dynamic adjustment of the state of the overflow port pool is achieved through the simulation model, and it is ensured that the pool runs within a safe range.

The invention discloses a side channel analysis method and device, equipment, a storage medium and a program product, and the method comprises the steps: obtaining a side channel segment related to polynomial multiplication of a secret key in mask-protected Kyber algorithm side channel data, and inputting the side channel segment into a Hamming weight classification model, the Hamming weight prediction value of the polynomial multiplication operation output result is obtained, and the Hamming weight classification model is obtained by adopting public stage data training of mask-protected Kyber algorithm side channel data; generating a candidate value set of different mask shares related to the same key coefficient according to the Hamming weight predicted value; screening out a target key coefficient from the candidate value set; and under the condition that the target key coefficient is successfully matched with the real key, determining that the side channel vulnerability that the key is stolen exists in the Kyber algorithm protected by the mask.

A method for writing data sets (DS) into a relational database (13), wherein each data set (DS) comprises at least two attributes (A), includes the steps of: determining at least one value for at least one of the attributes (A1) as criterion for filtering the data sets (DS) by the attribute value; computer implemented outputting data sets (DS) filtered by the attribute value in an output view (AS) and providing an input field (EF) in the output view (AS) for entering a value for another one of the attributes (A4); in response to entering a value in the input field (EF) of the output view (AS) computer implemented generating a complete data set (DS4) including the value set as filter criterion for the corresponding attribute (A1) and the entered value for the other attribute (A4); and storing the generated dataset (DS4) in the relational database (13).

The invention provides a data exception alarm method and device, electronic equipment and a storage medium, and the method comprises the steps: screening and removing data, which does not meet a preset measurement condition, in hydrological data, and calculating a relative change value between two adjacent data in the screened hydrological data to obtain a relative change value set, sorting the data in the relative change value set and determining a median, obtaining a first judgment condition representing that the hydrological data is abnormal fluctuation data based on the median and a preset first threshold value, and obtaining a second judgment condition representing that the hydrological data is abnormal fluctuation data based on the target deviation set and a preset target deviation threshold value; and according to the first judgment condition and the second judgment condition, determining whether the hydrological data is abnormal fluctuation data, and if the hydrological data is abnormal fluctuation data, generating first alarm information. Abnormal fluctuation data in hydrological data can be effectively identified and processed, and false alarms are reduced.

The invention relates to the technical field of data processing, and discloses a multi-source data difference attribution system and method based on tensor decomposition, and the method comprises the steps: obtaining an original engineering quantity list file, and then carrying out decoupling to generate a standardized list item text sequence and an initial value set, constructing unit prices to form a tetrahedral geometric model set, mapping the tetrahedral geometric model set to a holographic feature vector space, and generating a multi-dimensional list feature vector set; constructing a standard cost reference vector field, identifying a difference vector set and generating a visual cost difference topological data model; a natural language query instruction is responded, a target difference attribution tensor is locked, a logic verification chain is constructed, and an intelligent diagnosis report is output; according to the method, a computable logic distance is established in multi-source heterogeneous data by reversely reconstructing a high-dimensional price-rate feature space, so that the problem of attribution blocking caused by dimension collapse in traditional cost auditing is effectively solved, and accurate traceability and compliance automatic diagnosis of cost differences are realized.

The invention discloses a static binary code taint analysis method based on a propagation action scope, and relates to the field of static binary code taint analysis, and the static binary code taint analysis method comprises the following steps: extracting a program instruction set and a control flow structure based on a disassembling result of a target binary code; calculating a value set with a source of the binary code based on the control flow graph; on the basis of the control flow diagram and the value set with the source, executing cross-function stain propagation analysis, and identifying memory positions or registers influenced by pollution in each function and propagation action ranges of the memory positions or the registers; and extracting all instruction sets using the taint data based on the taint and the propagation action range thereof. According to the method, the false alarm rate of static binary taint analysis can be reduced and the instruction set involved in the taint analysis can be reduced without increasing the analysis overhead, so that the method has important significance in improving the instrumentation efficiency and the operation efficiency of dynamic taint analysis and improving the accuracy of protocol reversion, fuzzy testing and vulnerability mining based on the taint analysis.

The invention provides a formal verification method and device for state transition implementation codes and a program product, and the method comprises the steps: compiling a first state transition function implemented by a C language Switch Case statement to obtain a bit code file; obtaining a Cryptol file, wherein the Cryptol file comprises a second state transfer function which is realized by using a Cryptol programming language; acquiring a first input parameter set and a first output value set corresponding to the first state transfer function, and acquiring a second input parameter set and a second output value set corresponding to the second state transfer function; simulating state transition function symbolic execution based on the first output value set, the second output value set, the first input parameter set and the second input parameter set, converting the state transition function symbolic execution into a satisfiability model theory problem, and solving the satisfiability model theory problem through an external solver to obtain a verification result of a state transition implementation code; the problem that whether state transition is implemented correctly or not cannot be guaranteed only through a manual checking mode can be solved, and the state transition code implementation correctness is guaranteed.

The invention discloses a block compression method, device and equipment based on a block chain, and a readable storage medium. The method comprises the steps of obtaining M to-be-packaged transactions; one to-be-packaged transaction comprises transaction attribute values of N transaction attribute types; according to the transaction attribute type priorities corresponding to the N transaction attribute types and the M transaction attribute values corresponding to the N transaction attribute types, performing position sorting processing on the M to-be-packaged transactions to obtain M position sorted transactions; adding transaction attribute values with the same transaction attribute type in the M position sorting transactions to the same to-be-compressed attribute value set to obtain N to-be-compressed attribute value sets; performing incremental coding compression processing on the N to-be-compressed attribute value sets to obtain N attribute value incremental codes; and performing block encapsulation processing on the N attribute value increment codes to obtain a to-be-chained block. By adopting the method and the device, the storage space required by the block can be reduced, and the performance of the block chain system is improved.

The embodiment of the invention discloses a network service platform adjusting method and device, electronic equipment and a medium. A specific embodiment of the method comprises the following steps: acquiring a user data set and a user comment information set from different sources; performing user anomaly detection on the fused user data set to obtain an abnormal user information set; removing an abnormal user information set from the fused user data set to obtain a target user information set; performing user clustering processing on the target user information set to obtain a user class cluster set; performing user churn rate prediction on the user class cluster set to obtain a user churn rate numerical value set; screening user churn rate values greater than or equal to a preset user churn rate threshold to obtain a target user churn rate value set; performing demand prediction on the user comment information set to obtain a user demand information set; and dynamically adjusting the network service platform. According to the embodiment, the user data processing efficiency can be improved, the user demand can be accurately predicted, and the experience feeling of the user on the network service platform is improved.

The invention relates to the technical field of power systems, and discloses a power grid load prediction method and device and a related product. The method comprises the following steps: acquiring a first single-region load prediction result reported by a plurality of sub-regions in a to-be-predicted region; obtaining an index value set corresponding to each sub-region; according to each index value set, calculating to obtain weight data of each sub-region; and according to each first single-region load prediction result and each piece of weight data, calculating to obtain an overall load prediction result of the to-be-predicted region. According to the invention, the index value can reflect the characteristics of load prediction quality, influence degree, difficulty and the like of each sub-region. Therefore, the weight data of each sub-region is calculated according to the first single-region load prediction result and the index value set, and the proportion of the weight data in the overall load prediction can be reasonably distributed. And finally, combining the load prediction result of the first single area with the weight, so that the overall load prediction result of the to-be-predicted area can be calculated more accurately and reliably.

The invention relates to a trend analysis method suitable for multiple types of sensors in an underground coal mine, and belongs to the field of data analysis. The method comprises the following steps: S1, data preparation and normalization: acquiring original time sequence data of a selected sensor, and normalizing the original time sequence data to obtain a normalized value; s2, performing multi-scale division on a normalized value set: dividing the normalized value set by adopting a multi-time-scale sliding window method; s3, feature calculation: calculating a sample standard deviation of a long-term window and linear regression slopes in a short-term window and the long-term window; s4, state classification based on rules: distributing a state statei for each time point ti according to the feature vectors calculated in the step S3 and a classification engine; and S5, outputting and merging: merging the state classification sequences output in the step S4 into an event list. According to the method, the historical data set with the state mark is automatically generated, the industrial AI training data bottleneck is cracked, and advanced model development such as predictive maintenance is accelerated.

Provided in the embodiments of the present disclosure are a data compression method, a data decompression method, a computing device, a computer storage medium and a computer program product. The data compression method comprises: determining data to be compressed that has been subjected to character coding according to a unicode coding format; according to a byte order, reading character data from the data to be compressed, and determining a first dictionary or a second dictionary hit by the character data; searching for a coding sub-space corresponding to the first dictionary or the second dictionary, and determining coded values, in corresponding coding sub-spaces, of dictionary data hit by the character data, so as to obtain a coded-value set corresponding to the data to be compressed; and on the basis of the coded-value set and the first dictionary, obtaining compressed data.

According to the embodiment of the invention, a method, a device, equipment, a medium and a program product for model training are provided. The method comprises the steps of obtaining a to-be-trained target model, wherein the target model comprises a plurality of processing blocks; in the process of training the target model by using the training data set, for each processing block in the at least one processing block, executing a processing process corresponding to the processing block on the input of the processing block at least based on a weight set in a parameter value set of the processing block to obtain non-quantization output of the processing block; performing quantization operation on a first product of the weight set and an affine transformation matrix for the processing block by using a quantization function to obtain a quantization weight set; executing a processing process corresponding to the processing block on a second product of the input and the inverse matrix of the affine transformation matrix at least based on the quantization weight set to obtain quantization output of the processing block; and iteratively updating the parameter value set of the target model and the affine transformation matrix corresponding to the at least one processing block according to the optimization target.

This application provides a method and related equipment for processing printer model templates based on MD5 verification. The method includes: obtaining the local global template MD5 value from a local template database; sending the local global template MD5 value to a server for comparison with the cloud-based global template MD5 value on the server; if a global comparison result indicating inconsistency is received from the server, obtaining a set of local single template MD5 values ​​from the local template database; sending the set of local single template MD5 values ​​to the server for comparison with the cloud-based single template MD5 value set on the server; receiving templates for different printer models from the server; and updating the corresponding printer model templates in the local template database using the templates for the different printer models. This application can reduce network bandwidth consumption and shorten data processing time in the printer model template data synchronization scheme, thereby improving the update efficiency of printer model template data.

The invention discloses a knowledge base processing method based on a model, electronic equipment and an intelligent agent, and relates to the technical field of artificial intelligence, the intelligent agent can be embedded into an application program or an operating system to serve as a part of the application program or the operating system, and the intelligent agent is used for executing the following processing: obtaining input information; converting the input information into a first coding sequence; key values of all codes in the first code sequence are calculated in sequence to form a first key value set, and if a target mark exists in the input information, a second key value set of knowledge content of a target object indicated by the target mark is loaded to form a second key value set of knowledge content of the target object; the second key value set and the first key value set are combined to form a third key value set; reasoning a second coding sequence based on the third key value set; wherein the second key value set of the knowledge content of the target object comprises the key value of each code in the knowledge content stored before the input information is obtained.

This application provides a data transmission method and an apparatus. A first device determines a target lifting size based on a length of an information bit corresponding to to-be-transmitted data and an information column quantity of an LDPC base graph, where the target lifting size is in at least one of m lifting size sets, one lifting size set includes a lifting size corresponding to at least one of n bases, the m lifting size sets include lifting sizes corresponding to the n bases, a lifting size corresponding to an ith base ai in the n bases is determined based on a product of ai and an integer power of 2, n is greater than m, and ai is a positive integer. The first device determines, from m shifting value sets, a first shifting value set corresponding to the target lifting size.

The invention provides a solubility curve Tm value prediction method, and relates to the technical field of biological detection, and the method comprises the following steps: constructing a Tm value marking model and a Tm value prediction model; acquiring data information of a to-be-tested sample, and inputting the data information into the Tm value marking model; preprocessing the data information by a Tm value marking model and marking whether the to-be-detected samples have Tm values or not, and forming an effective sample set by the to-be-detected samples with the Tm values; extracting a plurality of peak values from the original dissolution curve to obtain an original Tm value set; encoding the original Tm value set according to the interval temperature; wherein the temperature bits in the original binary vector are connected and arranged according to interval temperatures; and inputting the original dissolution curve in the effective sample set into the Tm value prediction model to obtain a predicted Tm value. The temperature bits are connected and arranged according to the interval temperatures, so that the model learns the correlation between the adjacent temperatures, the influence of the temperature change on the Tm value is more accurately captured, and the prediction accuracy of the Tm value is improved.

This document describes techniques related to downlink transmission for high-speed scenarios. In one embodiment, a method includes: decoding a Transport Configuration Indicator (TCI) configuration element (IE) associated with a Physical Downlink Shared Channel (PDSCH) or Physical Downlink Control Channel (PDCCH), the TCI configuration IE including a first Tracking Reference Signal (TRS) configuration having a first Non-Zero Power Channel State Information Reference Signal (NZP-CSI-RS) resource identity (ID) and a Quasi-Co-address (QCL) type value set to type A, and a second TRS configuration having a second NZP-CSI-RS resource ID and a QCL type value set to type A; determining, based on the TCI configuration IE, that the Tracking Reference Signal (TRS) is associated with the PDSCH or PDCCH; determining parameters based on the TRS; and decoding the PDSCH or PDCCH based on the determined parameters. Other embodiments may be described and claimed.

The application relates to the field of semiconductor counting statistics, and discloses a DDR memory counting statistics method, equipment and a storage medium. The method comprises the following steps: when a counting request is received; performing polling allocation combination processing on a counting statistics ID and a counting request value to generate an allocation array; querying a memory statistics value corresponding to the counting statistics ID from a preset DDR memory; accumulating the memory statistics value and the counting request value to generate a new memory statistics value, and writing the new memory statistics value into the DDR memory; when a reading statistics request is received; querying all statistics values corresponding to the reading statistics ID from the DDR memory according to a reading statistics ID to generate a statistics value set, wherein the reading statistics ID is used to poll a Block counter value in a Block counter value set; and calculating a statistics output value corresponding to the reading statistics request based on the statistics value set. In the embodiment of the application, the characteristics of the DDR address are used for addition and subtraction of statistics values, a complex Cache mechanism is not needed, and the logical complexity and timing convergence difficulty are reduced.

To provide a method for effectively verifying a remitter or a recipient while technically protecting personal information, and an electronic device for performing the method.SOLUTION: Identifying, by the electronic device, a set of hash values of the sender's name generated based on the name information and a set of hash values of the sender's date of birth generated based on the date of birth information, and determining, by the electronic device, a S205 based on the set of hash values of the sender's name and the set of hash values of the sender's date of birth; A verification request message including a set of hash values of a name of a sender, a set of hash values of a birth date of the sender, and a receipt virtual asset address is transmitted to a receipt server, a response message including verification result information on whether the sender and the recipient match each other, which is performed based on the S210 and verification request message, is received from the receipt server, and it is determined whether to transfer virtual assets from a sending virtual asset address of the sender to the receipt virtual asset address based on the S220 and verification result information.SELECTED DRAWING: Figure 2a