923 results about "Coding block" patented technology

The invention provides a code retrieval method and device and related equipment, and relates to the technical field of retrieval. The method comprises the following steps: segmenting each source code file to obtain a plurality of code blocks of each source code file; determining description information of each code block, and determining an association relationship between any two code blocks; generating a multi-layer code knowledge graph based on the description information of each code block and the incidence relation between any two code blocks; obtaining query information of a user, and determining a first code block based on the query information; determining a second code block related to the first code block in the multi-layer code knowledge graph; and generating a retrieval result by using the first code block and the second code block. Through the technical means, the problem of low code retrieval accuracy in related technologies is solved.

The invention provides a script preservation method which can be applied to the technical field of artificial intelligence. The script preservation method comprises the following steps: carrying out semantic understanding on demand related files by utilizing a large model, extracting core elements, and generating a structured demand change list; analyzing to obtain an abstract syntax tree of an existing script code, and constructing a mapping relationship between the software automation test element and a script code block; dynamically comparing the structured demand change list with an existing script code structure tree based on a mapping relation between software automation test elements and script code blocks, identifying a script range influenced by demand change, and outputting a change influence matrix; and based on the change influence matrix, generating an automatic script code snippet needing to be updated, and realizing preservation of the automatic script. The invention further provides a script preservation device and equipment, a storage medium and a program product.

The invention provides an intelligent instruction execution method, system and device and a medium. The intelligent instruction execution method comprises the steps that current environment state information of a robot and a natural language task instruction input by a user are obtained; generating cue words at least based on the environment state information, the task instruction, a predefined standardized function interface document and a thinking chain example, and inputting the generated cue words into a large language model; the method comprises the following steps: decomposing a task instruction into sub-task sequences by utilizing a large language model, and organizing the decomposed sub-task sequences into complete executable code blocks; performing static logic verification on the code blocks before code execution; if verification is not passed, error description is generated and fed back to the large language model for code correction; and in a code execution stage, if abnormal information is returned after the code is executed, the abnormal information is returned to the large language model, so that the large language model regenerates a correction code according to an abnormal reason. The autonomous decision-making level and the task execution success rate of the robot are improved.

Disclosed are methods and apparatuses for decoding an image. A method includes receiving a bitstream obtained by encoding the image; dividing a first coding block into a plurality of second coding blocks; generating a prediction block of a second coding block based on syntax information obtained from the bitstream; and reconstructing the second coding block based on the prediction block and a residual block of the second coding block, the residual block being obtained by performing a dequantization and an inverse-transform on quantized transform coefficients from the bitstream. The first coding block has a recursive division structure. The first coding block is divided based on at least one of a quad tree division, a binary tree division or a triple tree division.

The invention discloses a version updating method and device, equipment and a storage medium, and relates to the technical field of version updating, and the version updating method comprises the following steps: obtaining to-be-updated version information of to-be-updated software, intercepting an original structured query language (SQL), and generating a syntax tree corresponding to the original SQL, the to-be-updated version information comprises to-be-updated SQL and to-be-updated code blocks; according to the to-be-updated SQL, performing field updating on the syntax tree and then converting to generate a target SQL; according to the code task identifier of the to-be-updated code block, determining a code position of an original code block, and generating a target code; based on the target SQL and the target code, version updating and testing are conducted on the software to be updated, and a version updating task is completed. According to the method, a low-code platform can be helped to accurately identify data and code change ranges, codes and scripts are automatically combined, and the purpose of reducing version updating time is achieved.

The invention provides a multi-agent-based code conversion system and method. The method comprises the following steps: preprocessing an input source code; performing block processing on the source code to obtain source code blocks, obtaining variable information and array size information of the blocks corresponding to the source code, and integrating the variable information and the array size information into complete variable information; fusing the code description file, the code blocks of the source codes and the variable information of the blocks corresponding to the source codes to generate target code blocks; debugging and checking the target code blocks in combination with the source code blocks; the code description file is used for guiding the intelligent agent to establish a jumping type relation in the code conversion process, and it is ensured that the generated target code better meets the actual function requirement of hardware; a block identifier is inserted into a long code to divide the long code into a plurality of blocks, so that the limitation on long code conversion is relieved; codes of different styles can be output by simply adjusting the cue word of the intelligent agent; through multiple times of debugging verification, the conversion quality of the target code is ensured.

Disclosed are methods and apparatuses for decoding an image. A method includes receiving a bitstream obtained by encoding the image; dividing a first coding block into a plurality of second coding blocks; generating a prediction block of a second coding block based on syntax information obtained from the bitstream; and reconstructing the second coding block based on the prediction block and a residual block of the second coding block, the residual block being obtained by performing a dequantization and an inverse-transform on quantized transform coefficients from the bitstream. The first coding block has a recursive division structure. The first coding block is divided based on at least one of a quad tree division, a binary tree division or a triple tree division.

The invention discloses a multi-modal sentiment analysis method and device based on feature decoupling and guide correction, and relates to the technical field of sentiment analysis, and the method comprises the steps: building an initial multi-modal sentiment analysis network, the initial multi-modal sentiment analysis network comprises a multi-modal feature extraction mapping block, a cross-modal interaction enhancement coding block, a feature decoupling block, a guide type gating semantic correction block and a low-rank fusion sentiment analysis block, and the total loss value of a training sentiment analysis result for training the multi-modal sentiment data is determined; and performing optimization iteration on the initial multi-mode sentiment analysis network according to the total loss value until a target multi-mode sentiment analysis network is determined, and outputting a target sentiment analysis result of the to-be-analyzed multi-mode sentiment data by adopting the target multi-mode sentiment analysis network. Based on the above scheme, the multi-modal fusion emotion recognition precision is improved.

The invention relates to a self-adaptive selection restoration method for defocus blurred image restoration, and belongs to the technical field of image restoration. The method comprises the following steps of: establishing a multi-scale branch which consists of three coding blocks and decoding blocks and is used for processing images with different resolutions; in each branch, shallow layer features of an input image with the corresponding resolution are extracted through a convolutional layer, and then image reconstruction is carried out from the shallow layer features by self-adaptive selection modules in corresponding coding blocks and decoding blocks; wherein each self-adaptive selection module comprises a self-adaptive double-branch fractional order module and a double-path fusion strategy based on gating reweighting, so that image reconstruction and feature fusion among different branches are carried out respectively, and finally a reconstructed image is output. Compared with the prior art, the method provided by the invention can realize a more efficient restoration effect on the compressed blurred image with lower model parameter quantity and calculation complexity.

Disclosed are methods and apparatuses for decoding an image. A method includes receiving a bitstream obtained by encoding the image; dividing a first coding block into a plurality of second coding blocks; generating a prediction block of a second coding block based on syntax information obtained from the bitstream; and reconstructing the second coding block based on the prediction block and a residual block of the second coding block, the residual block being obtained by performing a dequantization and an inverse-transform on quantized transform coefficients from the bitstream. The first coding block has a recursive division structure. The first coding block is divided based on at least one of a quad tree division, a binary tree division or a triple tree division.

The invention provides a cross-modal semantic alignment method, device and equipment based on fine-grained semantic decoupling, and the method comprises the steps: inputting a to-be-aligned image into a pre-training image encoder of a cross-modal semantic alignment model, and obtaining a global semantic feature, meanwhile, a semantic feature extraction module is used in a high-level encoding block of the image encoder and based on a preset query vector, high-level visual feature extraction is conducted, visual fine-grained semantic features are obtained and input into a semantic feature decoupling module to be decoupled, and fine-grained semantic features of attributes, objects and combinations are obtained; inputting prompt words corresponding to attributes, objects and combinations in the to-be-aligned image into a text encoder for text feature extraction to obtain soft prompt text features of the attributes, the objects and the combinations; and inputting the fine-grained semantic features and the soft prompt text features corresponding to the global semantic features, the attributes, the objects and the combinations into a loss calculation module for image-text alignment, and performing linear processing on an alignment result to obtain an alignment prediction result.

An image decoding / encoding apparatus and method are discussed. The image decoding method can include obtaining a coding block by partitioning an image, determining a split type for the coding block based on a size of the coding block, deriving a current coding block split from the coding block based on the split type for the coding block, and reconstructing the current coding block.

A method of decoding video data includes receiving, from bitstream, video data corresponding to a non-skip mode coded block; and determining a parameter set associated with the non-skip mode coded block. In accordance with a determination that the parameter set includes information that satisfies different conditions, the method further includes receiving a first syntax element, a second or a third syntax element from the bitstream, the first syntax element is received and the first syntax element has a first value, the third syntax element is received. The method further includes decoding the non-skip mode coded block based on palette mode when the first syntax element has a first value and the third syntax element has the first value; and decoding the non-skip mode coded block based on intra prediction mode when the first syntax element has the first value and the third syntax element has a second value.

Disclosed are methods and apparatuses for decoding an image. A method includes receiving a bitstream obtained by encoding the image; dividing a first coding block into a plurality of second coding blocks; generating a prediction block of a second coding block based on syntax information obtained from the bitstream; and reconstructing the second coding block based on the prediction block and a residual block of the second coding block, the residual block being obtained by performing a dequantization and an inverse-transform on quantized transform coefficients from the bitstream. The first coding block has a recursive division structure. The first coding block is divided based on at least one of a quad tree division, a binary tree division or a triple tree division.

A method and apparatus for video coding are disclosed. According to the method for the decoder side, encoded data associated with a current block comprising a first-colour block and a second-colour block are received. An inherited model parameter set is determined from a previously coded block coded in a first CCLM related mode, wherein the inherited model parameter set comprises a first scaling parameter associated with the first CCLM related mode. A final inherited model parameter set is derived if an update value for the inherited model parameter set is determined, where the final inherited model parameter set is determined based on the first scaling parameter and the update value. Then, the encoded data associated with the second-colour block are decoded using prediction data based on an updated CCLM related model associated with the final inherited model parameter set. A method and apparatus for the encoder side are also disclosed.

Aspects of the present disclosure relate to wireless communications, and more particularly, to techniques for multiple-input multiple-output (MIMO) communications One aspect provides a method for wireless communication. The method includes receiving a signal using a multiple-input multiple-output (MIMO) receiver, wherein the signal includes at least a first code block (CB) including a first part received via a first layer of the MIMO receiver and a second part received via a second layer of the MIMO receiver, wherein the second part of the first CB is shifted within a spectrum by at least two resource positions with respect to the first part of the first CB; performing interference cancellation for the first CB; and decoding the first CB after performing the interference cancellation.

A wireless device may receive a first code block of one or more code blocks associated with a transport block for the wireless device. The wireless device may output, using a neural network model associated with the wireless device, a set of low-density parity-check (LDPC) quantization values for a set of iterations of an LDPC decoding procedure for the first code block. In some examples, the set of LDPC quantization values may include respective LDPC quantization values for respective iterations of the set of iterations. The wireless device may perform one or more iterations of the LDPC decoding procedure for the first code block in accordance with one or more LDPC quantization values of the set of LDPC quantization values output using the neural network model.

This encoder comprises circuitry and a memory connected to the circuitry. The circuitry generates a plurality of feature maps by means of a neural network having one or more layers on the basis of an input image to be processed, generates a plurality of unit feature maps on the basis of the plurality of feature maps by packing a plurality of pixels included in at least one feature map into at least one encoded block, generates a picture by arranging a plurality of encoded blocks corresponding to the plurality of unit feature maps, and encodes the picture into a bitstream, and in the generation of the unit feature maps, the upper left boundary of each unit feature map is matched with the upper left boundary of any encoded block.

Provided is a terminal for performing transmission and reception of data in a wireless communication system, wherein the terminal receives, from a base station, first configuration information for code block group (CBG)-based uplink transmission, and second configuration information including a downlink feedback information (DFI) bitmap configuration method, transmits a CBG-based uplink signal to the base station, based on the first configuration information, receives, from the base station, DFI including feedback information about the CBG-based uplink signal, based on the second configuration information, and determines a contention window value of the terminal, based on the feedback information included in the received DFI, wherein the DFI bitmap configuration method indicates whether the feedback information is for feedback in units of transport blocks (TBs) or for feedback for one or more CBGs.

The invention relates to an abnormal code positioning method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring a graph representation vector and a code text representation vector corresponding to a to-be-detected source code; the graph representation vector is used for representing a code structure corresponding to the source code to be detected; the code text representation vector is used for representing code semantics corresponding to the source code to be detected; according to the graph representation vector and the code text representation vector, positioning an abnormal code block from the source code to be detected; determining a self-attention score corresponding to each code statement in the abnormal code block through a self-attention mechanism; the self-attention score is used for representing the influence degree of the corresponding code statement on the condition that the abnormal code block is predicted to be abnormal; and determining an abnormal code area of the abnormal code block according to the self-attention score corresponding to each code statement in the abnormal code block. By adopting the method, the abnormal code area positioning accuracy can be improved.

Block decoding in an artificial neural network is provided. An encoded block comprising a plurality of encoded values, each encoded value comprising a mantissa, is read. The encoded block's scaling information, which includes an exponent and a mantissa, is read. Each of the plurality of encoded values is decoded according to the scaling information to produce a plurality of decoded values.

Disclosed are methods and apparatuses for decoding an image. A method includes receiving a bitstream obtained by encoding the image; dividing a first coding block into a plurality of second coding blocks; generating a prediction block of a second coding block based on syntax information obtained from the bitstream; and reconstructing the second coding block based on the prediction block and a residual block of the second coding block, the residual block being obtained by performing a dequantization and an inverse-transform on quantized transform coefficients from the bitstream. The first coding block has a recursive division structure. The first coding block is divided based on at least one of a quad tree division, a binary tree division or a triple tree division.

This application discloses a code block processing method, apparatus, device, and readable storage medium, relating to the field of communication technology, to ensure the reliability of MTN networks. The method includes: if an MTN channel code block carrying an fgMTN channel is incorrectly marked, or the code block contains an invalid synchronization header, or the code block is a control code block with an invalid type field, or the code block is an erroneous code block, replacing the fgMTN channel code block carried within the MTN channel code block with the erroneous code block. Embodiments of this application can ensure the reliability of MTN networks.

In some implementations, a process implemented in a receiver device may include receiving a signal including one or more DMRS resource elements. The process may include synchronizing the received signal in the time domain, demodulating OFDM symbols of the signal, obtaining a first resource grid from the demodulated OFDM symbols, and performing multi-layer channel estimation utilizing the DMRS resource elements and the first resource grid to generate an estimate of a first channel matrix. Further, the process may include equalizing the first resource grid using the first channel matrix to obtain an equalized resource grid for each of a plurality of transmitted layers. In addition, the process may include de-interleaving, de-mapping, demodulation, and descrambling of the equalized resource grid for each of the plurality of transmitted layers to obtain LLR for a plurality of coded blocks, and channel decoding of the LLR to output a plurality of reassembled code-blocks.

The invention relates to the technical field of artificial intelligence, and particularly provides a model training method, a test case generation method, equipment, a medium and a product. According to the method, multiple natural supervision signals of software source codes are extracted, and multi-task instruction fine tuning training of a large-scale language model is realized based on the multiple natural supervision signals of the source codes, so that the large-scale language model can learn semantic features complementary with code features or code features complementary with the semantic features; therefore, when the test case is generated by utilizing the large-scale language model subjected to fine tuning training, the large-scale language model can accurately extract complementary semantic features or code features no matter whether the original data input into the large-scale language model is a code block or a demand description text; the purpose of jointly generating the test case in combination with the semantic features and the code features is achieved, the accuracy of the generated test case is improved, and the problem of feature missing in related technologies is solved.

Methods, systems, and devices for wireless communications are described. A user equipment (UE) may receive, via time-frequency resources using multiple spatial layers, a plurality of code block parts associated with a codeword, where code block parts corresponding to the same code blocks are each offset in different time-frequency resources, where one or more first code block parts are initializing code block parts associated with first decoding characteristics that are different than second decoding characteristics with which one or more second code block parts are associated. The UE may decode, based on the first decoding characteristics and the second decoding characteristics, the one or more first code block parts and the one or more second code block parts.

The invention provides a code optimization method and system based on a knowledge graph, and the method comprises the steps: carrying out the source code analysis based on a to-be-optimized code library, constructing an abstract syntax tree, and determining a to-be-optimized code snippet based on the code complexity of a corresponding code in the abstract syntax tree; determining a reconstruction mode corresponding to each to-be-optimized code snippet based on the to-be-optimized code snippet in combination with a pre-constructed code optimization knowledge graph, and constructing a context dependency graph of each to-be-optimized code snippet based on the abstract syntax tree and the code optimization knowledge graph corresponding to each to-be-optimized code snippet; performing sliding window decomposition based on the context dependency graph to obtain a plurality of mutually associated code blocks, and performing simplification and fusion on the code blocks to obtain a current context; and inputting the reconstruction mode corresponding to each to-be-optimized code snippet and the current context into the large language model for structured reconstruction to obtain a target reconstruction scheme output by the large language model. According to the method and the device, the code optimization effect is improved.

Disclosed are methods and apparatuses for decoding an image. A method includes receiving a bitstream obtained by encoding the image; dividing a first coding block into a plurality of second coding blocks; generating a prediction block of a second coding block based on syntax information obtained from the bitstream; and reconstructing the second coding block based on the prediction block and a residual block of the second coding block, the residual block being obtained by performing a dequantization and an inverse-transform on quantized transform coefficients from the bitstream. The first coding block has a recursive division structure. The first coding block is divided based on at least one of a quad tree division, a binary tree division or a triple tree division.

Provided is a method (300) and a successive interference cancellation (SIC) based multi-user uplink receiver (200) for multi-user uplink transmission. The method comprises receiving (302) signal of one or more multiple-access (MA) scheme waveforms by a plurality of antennas (202.1, . . . , 202.R) from one or more users. The method further comprises determining (304) one or more effective channel matrices corresponding to the plurality of antennas. Thereby, the method comprises performing (306) channel equalization for signal received in a corresponding antenna by an effective channel matrix. Furthermore, the method comprises combining (308) the channel equalized signal. Subsequently, the method comprises detecting (310) Correctly Decoded Code Blocks (CCBs) and Wrongly Decoded Code Blocks (WCBs). Upon detecting CCBs and WCBs, the method comprises performing (312) the SIC on received signals from one or more users until all WCBs are converted to CCBs or a maximum number of threshold iterations are completed.

The invention discloses a substrate management controller firmware upgrading method and electronic equipment, and relates to the technical field of servers, and the substrate management controller firmware upgrading method comprises the steps that current version firmware and to-be-upgraded target version firmware are divided into code blocks; according to the semantic similarity between the code blocks and a preset similarity threshold value, to-be-processed code block pairs with large semantic difference are screened out, difference analysis is conducted on the to-be-processed code block pairs, final difference data are determined, and a difference package is generated and sent to the substrate management controller for firmware updating; the technical problem that the server is unstable due to the fact that the firmware upgrading time of the substrate management controller is long is solved, and the technical effects that the firmware upgrading time of the substrate management controller is greatly shortened, and meanwhile stability in the upgrading process is guaranteed are achieved.