Tencent Technology (shenzhen) Company Limited

Tencent Technology Company Limited

Xiaoyu Yu

Lixin Zhang

Yuwei Wang

Yi Gao

Mengsha Zhou

Yongjie Li

Siyu Xiao

Libin Ren

Hua Li

Dewei Chen

Wei Mao

Heng Zhang

Jiongchao Lin

Jianlin Gao

Zhenzhen Xu

Feng Han

Bin Zhang

Jiaxin Li

Bo Zhang

Leteng Weng

Feng Chen

Qiang Lu

Keren Li

Yan Xiong

Jinming Zhang

Chen Gong

Pengfei Wang

Qianya Lin

Wenrui Zhang

Xiaobin Fang

Yu Meng

Yi Wu

Junbin Kuang

Xuan Ye

Jianping Zhu

Shaomian Yao

Sixin Gu

LiXin Zhang

Rui Tang

Tanhong Li

Haixia Rao

Duokai Huang

Boen Liang

Zeming Fan

Weizhao Li

Leming Fang

Guowei Xiang

Xinji Nie

Bei Yang

Zheng Dai

Wenxin Liao

A data processing method includes: acquiring a data sorting request for a data sequence to be sorted, and invoking C data bitonic sorting components in response to the data sorting request, C being a positive integer greater than 1; initiating B data bitonic sorting tasks according to the data sequence and the C data bitonic sorting components, B being a positive integer greater than 1, the B data bitonic sorting tasks being respectively associated with different data subsequences of B data subsequences, and the B data subsequences being generated based on the to-be-sorted data sequence; operating the C data bitonic sorting components in parallel according to the B data bitonic sorting tasks to obtain B data sorting subresults; and combining the B data sorting subresults based on the C data bitonic sorting components to obtain a data sorting result for the data sequence.

Data processing method and apparatus, computer device, and computer-readable storage medium

The present disclosure provides a systolic array-based data processing method that includes determining an input splice quantity for the systolic array based on a target input depth and a standard input depth, and determining an output splice quantity for the systolic array based on a target output depth and a standard output depth; inputting the input data matching the input splice quantity to an input buffer of the systolic array in batches, without overlaps in the input data, and processing, by the systolic array, the input data in the input buffer to generate output data corresponding to each piece of input data; and in accordance with a determination that a quantity of output data received by an output buffer of the systolic array from the systolic array matches the output splice quantity, outputting, in the output buffer, output data having a quantity matching the output splice quantity in batches.

Systolic array-based data processing method and apparatus, medium, and program product

A system parses a very long instruction word (VLIW) to obtain an execution parameter. The system obtains a first sliding window width count, a first sliding window height count, a first feature map width count, and a first feature map height count that correspond to first target data. In accordance with a determination that the first sliding window width count falls within the sliding window width range, the first sliding window height count falls within the sliding window height range, (the first feature map width count falls within the feature map width range, and the first feature map height count falls within the feature map height range, the system determines an offset of the first target data. The system also obtains a starting address of the first target data, and adds the starting address to the offset to obtain a first target address of the first target data.

Address generation method, related apparatus, and storage medium

This application discloses a pooling processing method, applied to a pooling processing system of a convolutional neural network. The method includes receiving a feature map comprising a width dimension and a height dimension; establishing a data region comprising k+m columns and n rows, k, m, and n being integers; transferring data blocks each of size k in width and size n in height from the feature map in a width-height sequential scanning order into columns m+1 to k+m and rows 1 to n of the data region in pooling cycles; and in each pooling cycle corresponding to a data block transferred from the feature map to the data region: retaining columns k+1 to k+m of the data region of a previous pooling cycle to columns 1 to m of the data region; performing a first level pooling within each column of the data region to obtain k+m first level pooling values; and performing P second level pooling values by pooling P sets of first level pooling values selected according to a position of the data block relative to a width boundary in the feature map, P being an integer.

Pooling processing method and system applied to convolutional neural network

This application discloses a pooling processing method, applied to a pooling processing system of a convolutional neural network. The pooling processing system includes a first storage device, a data region, a pooling computation kernel, and a pooling controller. The method includes: reading, by the pooling controller, k pieces of feature data from the first storage device in each reading cycle, the k pieces of feature data being components in a feature map generated by a convolution operation of the convolutional neural network, and k being an integer greater than 1; writing, by the pooling controller, the k pieces of feature data read from the first storage device into the data region, wherein the k pieces of feature data form one group among n groups of k pieces of data with each group arranged in a first dimension and the n groups arranged in a second dimension, wherein the n groups of k pieces of data are written into the data region in an updating cycle, wherein a duration of the updating cycle is n times a duration of the reading cycle, and wherein n cis an integer greater than 1; and transmitting, after the updating cycle is ended, data in the data region to the pooling computation kernel to perform a pooling operation, wherein the data in the data region comprises the n groups of k pieces of data and last m groups of data from a previous updating cycle with each group along the second dimension, wherein the last m groups of data are temporarily stored in the data region for use in pooling calculation by the pooling computation kernel in a next updating cycle. The technical solution in this application reduces the number of storage, numbers of reading and writing due to data reuses, and improves the efficiency of pooling processing.

The present disclosure describes methods, devices, and storage mediums for adjusting computing resource. The method includes obtaining an expected pooling time of a target pooling layer and a to-be-processed data volume of the target pooling layer; obtaining a current clock frequency corresponding to at least one computing resource unit used for pooling; determining a target clock frequency according to the expected pooling time of the target pooling layer and the to-be-processed data volume of the target pooling layer; and in response to that the convolution layer associated with the target pooling layer completes convolution and the current clock frequency is different from the target clock frequency, switching the current clock frequency of the at least one computing resource unit to the target clock frequency, and performing pooling in the target pooling layer based on the at least one computing resource unit having the target clock frequency.

Neural network operational method and apparatus, and related device

Embodiments of this application relate to a convolutional computing accelerator, a convolutional computing method, and a convolutional computing device, which belong to the technical field of electronic circuits. The convolutional computing accelerator includes: a controller, a computing matrix, and a first cache. The computing matrix comprising at least one row of computing units, each row of computing units comprising at least two adjacent connected computing units. The controller is configured to control input data of each row of computing units to be loaded into the first cache, and to control the input data loaded into the first cache to be inputted into the two adjacent computing units in a corresponding row. Each of the computing units in the corresponding row is configured to perform, in a first clock cycle, a convolutional computation based on received input data and a pre-stored convolutional kernel.

Convolutional computing accelerator, convolutional computing method, and computer-readable storage medium

Embodiments of this application disclose an instruction set processing method based on a chip architecture and apparatus, and a computer-readable storage medium. The method includes compiling a deep learning model based on the architecture of the chip, to obtain a deep learning instruction set corresponding to the chip; compressing the deep learning instruction set, to obtain a compressed instruction set; and storing the compressed instruction set in an instruction set buffer of the chip by writing in a register, the compressed instructions executing a task.

Instruction set processing method based on a chip architecture and apparatus, and storage medium

After a user logs in to a client, a first request of the user is sent to a server, and after the first request is authenticated, a communication connection between the client and the server is established; a system message sent by the server is received in a user login interface to which the user has logged in; the system message is generated by the server to contain at least text-format information capable of being displayed at the client.

Text and custom format information processing method, client, server, and computer-readable storage medium

An information processing method, which is applied to a server, includes: a public message is transmitted to a first client, the public message being a message issued to the first client by the server, the public message carrying first information and a second information set including N pieces of second information, with N being a positive integer, the first information is information only allowed to be gotten by the first client and the second information is information only allowed to be gotten by a client in a sharing relationship with the first client; and a request for getting the second information is received from a second client, it is verified whether the second client is in the sharing relationship with the first client, and after the second client is determined to be in the sharing relationship with the first client, the second client is allowed to get the second information.

Information processing method, client, server and computer-readable storage medium

An association logic that associates a non-text message type with specified information is established. The association logic includes at least identification of the non-text message type, allowing a message type in line with the association logic to be identified by the identification. A first message is monitored. A first identification corresponding to the first message is obtained by analysing the first message. It is detected, according to the first identification, whether the first message is of the message type in line with the association logic. It is determined that the first message supports display of the specified information when it is detected that the first message is of the message type in line with the association logic.

Information processing method, terminal, and computer-readable storage medium

A first selectable command is displayed on a User Interface (UI) of the UE, and a first request is sent to a server via the UI, such that the server returns a first message of a non-text type; the first message returned by the server is received, and it is determined that the first message supports display of specified information; first operation applied on a result displayed by the first message is received, and a second request is sent, via the first operation, to the server to draw random information; in response to the first operation, display of the specified information is triggered, and the random information drawn from the server is received.

Information processing method, user equipment, server, and computer-readable storage medium

Growing community of inventors

Shenzhen, China

Xiaoyu Yu