Abstract: Provided are a LED display unit group and a display panel. The LED display unit group includes a circuit board and pixel units arranged in an array of m rows and n columns on the circuit board. The circuit board includes N metal line layers stacked in sequence and an insulating plate located between adjacent metal line layers. The N metal line layers are electrically connected through a conductive via on the insulating plate, where N?2. Each pixel unit includes at least two LED light-emitting chips with different light-emitting colors, where m?2, n?2. The first metal line layer includes m common A-electrode pads, multiple A-electrode pads and multiple B-electrode pads.

Abstract: A method for generating a road annotation, a device, and a storage medium are provided. The method may include: generating a road quantity and a road width in a tag picture; generating, for each road in the tag picture, a start point and an end point of the road; generating at least one point between the start point and the end point; drawing, for two adjacent points, a line segment from a previous point to a next point, where a width of the line segment is equal to the road width; and generating slanted box annotation information based on a coordinate of the previous point and a coordinate of the next point, where the slanted box annotation information includes an intersection point of diagonal lines, a width, a height and a slant angle of a slanted box.

Abstract: A fold testing device, including a support mechanism, an adjusting mechanism, and an attaching mechanism. The support mechanism includes a supporting body, a first position limiting part, and a second position limiting part. The first position limiting part and the second position limiting part are arranged on a side of the supporting body along a first direction. The supporting body, the first position limiting part and the second position limiting part are enclosed to form a positioning part. A flexible display module is received in the positioning part. The adjusting mechanism is installed on the supporting body, and configured to drive at least one of the first position limiting part and the second position limiting part to move along the first direction. The attaching mechanism includes an elastic attaching body. The elastic attaching body is movable toward the supporting body along a second direction. The elastic attaching body engages with the positioning part.

Abstract: Introduced here is at least one technique to better estimate interference at a receiver. The technique includes receiving a plurality of reference signals, which each have information indicative of noise. Thus, the technique further includes, for each reference signal, determining a noise estimation and determining a distance metric and log-likelihood ratio (LLR) of the noise estimation. Once the distance metric and LLR of each reference signal is determined, the receiver can determine a final LLR based on the distance metric and LLR of each reference signal. In this manner, a final LLR is determined. This technique can be applied by any device operating on MIMO technology.

Abstract: A data processing method, an apparatus, and a device. The data processing method may be performed by a first communication device, and the first communication device is a transmit end of encoded data. When simultaneously sending one or more data streams, the first communication device places information bits in a data stream in a specified order. A plurality of data streams may be arranged in a specified order, so that joint encoding without feeding back information can be implemented, and transmission close to a channel capacity can be implemented.

Abstract: According to one aspect of the present disclosure, a baseband chip may be configured to receive a data stream associated with a channel The baseband chip may be configured to select a first anchor point from the first constellation points of the first transmission layer. The baseband chip may be configured to select a first subset of constellation points from the first constellation points and the second constellation points. The baseband chip may be configured to perform a first iteration of a recursive tree search. The baseband chip may be configured to determine a first path metric based at least in part on the first iteration of the recursive tree search operation. The baseband chip may be configured to select a second anchor point from the second constellation points of the second transmission layer. The second anchor point may be associated with a second iteration of the recursive tree search.

Abstract: A method and an apparatus for encoding a polar code concatenated with a cyclic redundancy check (CRC), where M bits are selected from K bits in the sequence to perform CRC encoding. The M bits are determined based on reliability of K polarized subchannels on which the K bits are placed and/or row weights of K rows, in a first matrix, corresponding to the K polarized subchannels on which the K bits are placed. The first matrix is an encoding matrix of polar encoding. Polar encoding is performed on the K bits and obtained CRC check bits. An encoded codeword is output.

Abstract: A method of training an image-text retrieval model, a method of multimodal image retrieval, an electronic device and a storage medium, each relating to the technical field of artificial intelligence, and in particular, to fields of computer vision and deep learning technologies. Sample data including a sample text and a sample image is acquired. The sample text includes a sample text in a first language and a sample text in a second language. The sample text in the first language and the sample text in the second language are processed by using the text encoding sub-model to obtain a sample text feature of the sample data. The sample image is processed by using the image encoding sub-model to obtain a sample image feature of the sample data. The image-text retrieval model is trained according to the sample text feature and the sample image feature.

Abstract: A partial discharge (PD) detection apparatus for gas-insulated equipment includes a photon collector, an optical splitter, a first photoelectric conversion module, an ultraviolet fluorescent crystal, a second photoelectric conversion module, and a signal processing module, where the ultraviolet fluorescent crystal is configured to convert a second optical radiation signal into an optical radiation signal of an ultraviolet fluorescence band, and the signal processing module is configured to calculate first apparent intensity based on a first voltage signal output by the first photoelectric conversion module, calculate second apparent intensity based on a second voltage signal output by the second photoelectric conversion module, and determine discharge intensity of the optical radiation based on a ratio of the second apparent intensity to the first apparent intensity. Technical solutions of the present disclosure are not affected by an unknown distance between a discharge position and a detection point.

Abstract: A method for generating a sample image includes: obtaining an initial image size of an initial image; obtaining a plurality of reference images by processing the initial image based on different reference processing modes; obtaining an image to be processed by fusing the plurality of reference images; and determining a target sample image from images to be processed based on the initial image size.

Abstract: This application discloses a polar code encoding method and apparatus, which can improve encoding performance through codeword construction. The method includes: obtaining K information bits to be encoded, where K is a positive integer; determining a first bit sequence based on the K information bits to be encoded, where a length of the first bit sequence is N, the first bit sequence includes fixed bits and the K information bits to be encoded, and the K information bits to be encoded may further include a check bit; determining a second bit sequence based on the first bit sequence and an upper triangular matrix, where the upper triangular matrix is a matrix having N rows and N columns, and the upper triangular matrix may be an upper triangular Toeplitz matrix; and performing polar code encoding on the second bit sequence to obtain and output an encoded sequence.

Abstract: System and method for training a federated learning model asynchronously. The system includes a coordinator, an active computing device and a passive computing device in communication with each other. The active computing device has a processor and a storage device storing computer executable code. The computer executable code is configured to: train the federated learning model in the active computing device using dimensions of an instance in the active computing device; and instruct the at least one passive computing device to train the federated learning model in the at least one passive computing device using dimensions of the instance in the at least one passive computing device. The training instances in the active and the at least one passive computing devices do not correspond to each other at the same training time.

Abstract: A method for processing an uplink reference signal includes determining a current baseband processing payload, and processing each of N received uplink reference signals according to a preset first condition in response to the baseband processing payload being less than a first preset value or processing each uplink reference signal in a signal set according to a preset second condition in response to the baseband processing payload being greater than a second preset value. The second preset value is greater than or equal to the first preset value. The signal set includes one or more of the N uplink reference signals. A baseband processing payload corresponding to processing an uplink reference signal according to the preset second condition is less than a baseband processing payload corresponding to processing an uplink reference signal according to the preset first condition. N is a positive integer.

Abstract: The present invention relates to a mannanase PMan5A mutant having improved heat resistance, a gene encoding the mutant and application thereof. The mutant is obtained by a substitution of histidine with tyrosine at amino acid residue 93, phenylalanine with tyrosine at amino acid residue 94, leucine with histidine at amino acid residue 356, and/or alanine with proline at amino acid residue 389. The thermal tolerance of the single site mutation mutant H93Y, L356H and A389P are greatly improved over that of the wild-type mannanase PMan5A, and the thermal tolerance of the combination mutants shows the stack effect of the single site mutation, demonstrating the amino acids at the sites of 93, 94, 356, and 389 play an important role for the thermal stability of the mannanase of GH5 family.

Abstract: A display panel bonding device and a correction apparatus and method thereof. The correction apparatus includes: a fixture assembly including a first fixture and a second fixture coordinating with each other. The first fixture is provided with a first alignment mark, the second fixture is provided with a second alignment mark, and alignment between the first fixture and the second fixture is achieved via the first alignment mark and the second alignment mark; and an image acquisition assembly configured to acquire a first image of the first alignment mark of the first fixture and a second image of the second alignment mark of the second fixture respectively. The first image and the second image are used for acquiring an alignment parameter of the first alignment mark and the second alignment mark to correct alignment between the first fixture and the second fixture.

Abstract: The present disclosure provides example coding method, apparatus, and computer-readable storage medium. One example method includes sorting a plurality of pieces of run-length encoding (RLE) data based on their distribution probabilities. The plurality of pieces of sorted RLE data are mapped onto a plurality of pieces of reassembled data. The plurality of pieces of reassembled data are expanded into a plurality of pieces of binary data. A first matrix is generated based on the plurality of pieces of binary data. A code sequence is determined by processing a source signal, where the code sequence comprises a first set and a second set. A first vector is decoded to output a third set, where the third set includes one or more bits reserved after processing the source signal and one or more locations at which a decoding error occurs in one or more bits discarded after processing the source signal.

Abstract: A moving state reporting method includes communicating, by a terminal, with a network device using a beam to carry a signal reporting, by the terminal, moving state information of the terminal to the network device, where the moving state information includes a speed flag, and the speed flag indicates that a speed of the terminal exceeds a speed threshold, and receiving, by the terminal, beam management information from the network device.

Abstract: The invention provides pharmaceutical compositions and methods of use thereof for preventing or ameliorating disorders of the nervous system. More specifically, the invention provides pharmaceutical compositions, including phosphopeptides that when administered disrupt the physical interaction of TrkB with its signaling effector, phospholipase C?1 (PLC?1). The invention further provides method of treatment comprising administering phosphopeptides that uncouple TrkB from PLC?1 in order to prevent and/or ameliorate nervous system disorders such as epilepsy, anxiety, and seizures.

Abstract: System and method for prediction using a machine learning model. The system includes a coordinator, an active computing device and a passive computing device in communication with each other. The active computing device has a processor and a storage device storing computer executable code. The computer executable code is configured to: obtain parameters of the machine learning model; retrieve an instance from the local data; sample a random direction of the instance; compute a dot product of the random direction and the instance, and calculate a random feature; compute predicted values of the instance in the active and passive computing devices and summarize them to obtain a final predicted value; determine a model coefficient using the random feature, the final predicted value, and a target value of the instance; update the machine learning model using the model coefficient; and predict a value for a new instance.

Abstract: System and method for training a federated learning model asynchronously. The system includes a coordinator, an active computing device and a passive computing device in communication with each other. The active computing device has a processor and a storage device storing computer executable code. The computer executable code is configured to: train the federated learning model in the active computing device using dimensions of an instance in the active computing device; and instruct the at least one passive computing device to train the federated learning model in the at least one passive computing device using dimensions of the instance in the at least one passive computing device. The training instances in the active and the at least one passive computing devices do not correspond to each other at the same training time.