Suzhou Metabrain Intelligent Technology Co., Ltd.

Inspur Suzhou Intelligent Technology Co., Ltd

Rengang Li

Yaqian Zhao

Baoyu Fan

Zhenhua Guo

Li Wang

Hongwei Kan

Yanwei Wang

Ruyang Li

Gang Dong

Hongbin Yang

Haiwei Liu

Dongdong Jiang

Xiaochuan Li

Wei Liu

Xin Zhang

Jiangwei Wang

Hongliang Wang

Deshan Zhang

Rui Hao

Lingyan Liang

Jiaheng Fan

Weifeng Gong

Nan Wu

Le Yang

Dongdong Su

Xiangyi Deng

Kefeng Zhu

Dengwei Li

Qi Mu

Yaqiang Zhang

Xiaomin Zhu

Hui Wei

Chen Li

Jinzhe Jiang

Provided are a communication method and system for a distributed heterogeneous acceleration platform, a device and a medium. The method includes: after starting a collaborative acceleration task, determining, by a first target heterogeneous acceleration card in a distributed heterogeneous acceleration platform, a second target heterogeneous acceleration card from the distributed heterogeneous acceleration platform by querying an information table corresponding to the collaborative acceleration task; generating, by the first target heterogeneous acceleration card, a target data packet according to a predefined data packet format, and sending the target data packet to the second target heterogeneous acceleration card via a PCIE interface; and parsing, by the second target heterogeneous acceleration card, the target data packet according to the data packet format, and executing a corresponding read operation or write operation according to a parsing result, so as to complete the collaborative acceleration task.

Communication method and system for distributed heterogeneous acceleration platform, device and medium

A data processing method, a data processing apparatus, a distributed data flow programming framework, an electronic device, and a storage medium. The data processing method includes: dividing a data processing task into a plurality of data processing subtasks (S); determining, in a Field Programmable Gate Array (FPGA) accelerator side, a target FPGA acceleration board corresponding to each of the data processing subtasks (S); and sending data to be computed to the target FPGA acceleration board, and executing the corresponding data processing subtask by use of each of the target FPGA acceleration boards to obtain a data processing result (S). According to the method, a physical limitation of host interfaces on the number of FPGA acceleration boards in an FPGA accelerator side may be avoided, thereby improving the data processing efficiency.

Data processing method and apparatus, distributed data flow programming framework, and related component

A natural language processing method and apparatus, a device, and a readable storage medium, where the method includes: obtaining a target sentence to be processed, and determining each entity in the target sentence (S); for each entity in the target sentence, in response to the entity being present in a preset entity set, determining extended information for the entity, and adding the determined extended information after a location of the entity in the target sentence, to obtain an updated target sentence (S); and inputting the updated target sentence to a bidirectional encoder representations from transformer (BERT) model, such that the BERT model performs a natural language processing task, where in a process in which the BERT model performs the natural language processing task, an attention score between extended information of any entity in the target sentence and another entity in the target sentence is adjusted to zero (S).

Natural language processing method and apparatus, device, and readable storage medium using a three dimensional tensor to determine a relation between entities to aid in a question-answering task

The present application relates to the technical field of artificial intelligence, and discloses a visual positioning method and apparatus, a device, and a medium. The method includes: performing feature splicing on an image encoding feature and a text encoding feature; performing feature fusion on spliced encoding features to obtain a first fused encoding feature; performing noise correction on the first fused encoding feature and the text encoding feature on the basis of a preset cross-attention mechanism to obtain a corrected fused feature and a corrected text encoding feature, and performing feature fusion on the spliced encoding feature and the corrected text encoding feature to obtain a second fused encoding feature; and correcting a preset frame feature using a target encoding feature on the basis of the corrected fused feature and the second fused encoding feature to predict a regional position coordinate of a target visual object.

Visual positioning method and apparatus, device, and medium

Embodiments of the present application disclose a natural language processing method and apparatus, a device, and a readable storage medium. The method includes: obtaining a target sentence to be processed, and determining each first entity in the target sentence; for each first entity, in response to the first entity being present in a preset entity set, determining, in the preset entity set, a second entity in maximum correlation with the first entity, generating extended information based on the determined second entity, and adding the extended information after a location of the first entity in the target sentence, to obtain an updated target sentence, where the second entity is any entity in the preset entity set other than the first entity; and inputting the updated target sentence to a bidirectional encoder representations from transformer (BERT) model, such that the BERT model performs a natural language processing task.

Natural language processing method and apparatus, device, and readable storage medium

Provided is a direct memory access architecture, including: a direct memory access control component, a read data moving component, a write data moving component and a data storage component; wherein the direct memory access control component includes a control register, a read descriptor storage component, a write descriptor storage component, a read command transfer component and a write command transfer component; the control register is configured to obtain descriptor address information; the read descriptor storage component is configured to store a read descriptor obtained by using the control register; the write descriptor storage component is configured to store a write descriptor obtained by using the control register; the read command transfer component is configured to send, to the read data moving component, a read command; the write command transfer component is configured to send, to the write data moving component, a write command.

Direct memory access architecture, system and method, electronic device, and medium

Disclosed in the present disclosure are an image-text mutual retrieval model training method and apparatus, an image-text mutual retrieval method, and a device, applied to the technical field of retrieval. The image-text mutual retrieval model training method includes: acquiring training data pairs; inputting the training data pair in an initial model, and extracting text coding features of text training data and image coding features of image training data by using a text coding module and an image coding module in the initial model, respectively; calculating a training loss based on the text coding features and the image coding features, and performing parameter adjustment on the initial model based on the training loss; and in response to the training loss meeting a convergence condition, determining the initial model after the parameter adjustment as an image-text mutual retrieval model.

Method and apparatus for training an image-text mutual retrieval model, image-text mutual retrieval method, and device

Provided in the present application are a disk processing method and system, and an electronic device. The method comprises: when disk alarm information is detected, marking a corresponding alarm disk as a faulty disk; detecting the state of a disk group corresponding to the faulty disk; if the state of the disk group is degraded, marking the faulty disk as an isolation disk, and generating alarm information; if the state of the disk group is healthy, determining whether there is a redundant disk group; if there is no redundant disk group, operating the faulty disk according to a first preset rule, and generating alarm information; and if there is a redundant disk group, detecting the state of the redundant disk group, if the state of the redundant disk group is healthy, marking the faulty disk as the isolation disk, and generating alarm information, otherwise, operating the faulty disk according to a second preset rule, and generating alarm information.

Pedestrian re-identification method, system and device, and computer-readable storage medium

The present application discloses a password protection method. Aiming at the current problem that, when a plurality of passwords are used, the plurality of passwords may be invalidated after one password is compromised, a password protection method is provided, which comprises: after password authentication is successful, determining, according to a password currently passing the authentication, a specified bit of password invalidation enable identifiers stored in a storage medium, wherein there is only one specified bit because there is only one authenticated password; setting the specified bit to be in an allowed state and other bits to be in a non-allowed state, wherein the password invalidation enable identifiers are in one-to-one correspondence with passwords according to bits, and only when one bit of the password invalidation enable identifiers is in the allowed state, an invalidation operation on the passwords is allowed; determining whether a password invalidation request is received; and if a password invalidation request is received, performing invalidation operation on the password corresponding to the specified bit.

Password protection method and apparatus, and medium

A difference description statement generation method includes: performing feature concatenation on an image encoding feature and a text encoding feature; inputting a concatenated encoding feature to a preset image-text alignment unit constructed based on a preset self-attention mechanism to obtain a concatenated alignment feature; processing, by using a preset noise monitoring unit constructed based on the preset self-attention mechanism and a preset cross-attention mechanism, the text encoding feature, and an image alignment feature and a text alignment feature that are obtained by splitting the concatenated alignment feature respectively, to extract a difference signal; and generating a difference description statement based on the difference signal by using a preset difference description generation algorithm.

Difference description statement generation method and apparatus, device and medium

A heterogeneous acceleration method, including the following steps: analyzing a received target data packet to obtain data source end identification information; when the data source end identification information does not exist in a configuration information table, customizing a data message format according to the target data packet to obtain a target data message format, or when the data source end identification information exists in the configuration information table, reading the target data message format from the configuration information table; and according to the target data message format, performing acceleration processing on the target data packet to obtain processed data, and returning the processed data to a data source end.

Heterogeneous acceleration method, apparatus and system and computer readable storage medium

A high-resolution image matching method and system are provided. The method includes performing regional fidelity down-sampling on an initial high-resolution image to obtain a multi-level low-resolution image, performing local matching on the obtained multi-level low-resolution images using a method with global probes to obtain a matching result of the low-resolution images, and performing reverse refinement on the obtained matching result of the low-resolution image using overall consistency of the image matching, to obtain the matching results of the high-resolution images at all levels until the matching results of the initial resolution images are obtained, so as to reduce the computational complexity of the image matching process and improve the accuracy of the matching result, and then the matching result of the high-resolution image is obtained through reverse refinement based on the overall consistency.

High-resolution image matching method and system

A task scheduling method includes: when a task requirement is obtained, splitting the task requirement to obtain the plurality of subtasks having a constraint relationship; performing execution condition detection on non-candidate subtasks, determining a non-candidate subtask that satisfies an execution condition as a candidate subtask, and putting the candidate subtask into a task queue; performing state detection on a server network composed of edge servers to obtain server state information and communication information; inputting the server state information, the communication information, and queue information corresponding to the task queue into an action value evaluation model to obtain the plurality of evaluated values respectively corresponding to the plurality of scheduling actions; and determining a target scheduling action from the plurality of scheduling actions by using the evaluated values, and scheduling the candidate subtask in the task queue on the basis of the target scheduling action.

Task scheduling method and apparatus, electronic device, and readable storage medium

The present application discloses an automatic driving control method. In the method, parameters are optimally set by using a noisy and noiseless dual-strategy network, identical vehicle traffic environment state information is input into the noisy and noiseless dual-strategy network, a motion space perturbation threshold is set by using a noiseless strategy network as a comparison and a benchmark so as to adaptively adjust noise parameters, and motion noise is indirectly added by adaptively injecting noise into a strategy network parameter space, such that exploration of an environment and a motion space by a deep reinforcement learning algorithm may be effectively improved, automatic driving exploration performance and stability based on deep reinforcement learning is improved, and full consideration of influence of an environment state and driving strategies in vehicle decision-making and motion selection is ensured, thereby improving the stability and safety of an automatic vehicle.

Autonomous driving control method, apparatus and device, and readable storage medium

An image recognition method and apparatus based on a systolic array, and a medium are disclosed. The method includes: converting obtained image feature information into a one-dimensional feature vector; converting an obtained weight matrix into a one-dimensional weight vector, and allocating a corresponding weight group to each node in a trained three-dimensional systolic array model; performing multiply-accumulate of the feature vector and a weight value on the one-dimensional feature vector in parallel by using the three-dimensional systolic array model, to obtain a feature value corresponding to each node, the feature values with different values reflecting article categories contained in an image; and determining an article category contained in the image according to the feature value corresponding to each node and a pre-established corresponding relationship between the feature value and the article category.

Growing community of inventors

Jiangsu, China

Rengang Li