Abstract: A data feature extraction method and apparatus in the field of artificial intelligence are provided. An addition convolution operation is performed to extract a target feature in quantized data based on quantized feature extraction parameters, that is, to calculate a sum of absolute values of differences between the quantized feature extraction parameters and the quantized data, to obtain the target feature based on the sum. In addition, feature extraction parameters and data are quantized by using a same quantization parameter. According to this application, a storage resource is saved, a computing resource is saved, thereby reducing a limitation on an application of artificial intelligence to a resource-limited device. Further, when the extracted feature data is dequantized, the feature data may be dequantized based on the quantization parameters.

Abstract: A neural network training method in the artificial intelligence field includes: inputting training data into a neural network; determining a first input space of a second target layer in the neural network based on a first output space of a first target layer in the neural network; and inputting a feature vector in the first input space into the second target layer, where a capability of fitting random noise by the neural network when the feature vector in the first input space is input into the second target layer is lower than a capability of fitting the random noise by using an output space that is in the neural network and that exists when a feature vector in the first output space is input into the second target layer. This application helps avoid an overfitting phenomenon that occurs when the neural network processes an image, text, or speech.

Abstract: An electrical connector includes a conductive housing having a first receiving portion with an open end, a movable terminal having a first end inserted into the first receiving portion via the open end, the movable terminal having a second receiving portion, and a first elastic piece located in a space defined by the first receiving portion and the second receiving portion. A second end of the movable terminal opposite to the first end protrudes movably out of the first receiving portion against an elasticity of the first elastic piece. The first end of the movable terminal has a first elastic arm elastically abutting an inner side wall of the first receiving portion. The movable terminal is electrically connected to the conductive housing by each of the first elastic piece and the first elastic arm.

Abstract: This application provides a neural network model compression method in the field of artificial intelligence. The method includes: obtaining, by a server, a first neural network model and training data of the first neural network that are uploaded by user equipment; obtaining a PU classifier based on the training data of the first neural network and unlabeled data stored in the server; selecting, by using the PU classifier, extended data from the unlabeled data stored in the server, where the extended data has a property and distribution similar to a property and distribution of the training data of the first neural network model; and training a second neural network model by using a knowledge distillation (KD) method based on the extended data, where the first neural network model is used as a teacher network model and the second neural network model is used as a student network model.

Abstract: An electrical connector includes an electrical connector housing having a body with a receiving portion, a movable pin, a part of the movable pin is movably disposed in the receiving portion, a first elastic piece disposed in the receiving portion, and a second elastic piece located between a side wall of the receiving portion and the movable pin and elastically abutted against an outer circumferential surface of the movable pin. A first end of the movable pin movably protrudes out of the receiving portion against an elasticity of the first elastic piece. The movable pin is electrically connected to the electrical connector housing by the first elastic piece and the second elastic piece.

Abstract: This application relates to an image recognition technology in the field of computer vision of artificial intelligence, and provides an image classification method and apparatus. The method includes: obtaining an input feature map of a to-be-processed image; performing feature extraction processing on the input feature map based on a feature extraction kernel of a neural network, to obtain an output feature map, where each of a plurality of output sub-feature maps is determined based on the corresponding input sub-feature map and the feature extraction kernel, at least one of the output sub-feature maps is determined based on a target matrix obtained after an absolute value is taken, and a difference between the target matrix and the input sub-feature map corresponding to the target matrix is the feature extraction kernel; and classifying the to-be-processed image based on the output feature map, to obtain a classification result of the to-be-processed image.

Abstract: This application relates to an image recognition technology in the field of computer vision in the field of artificial intelligence, and provides an image classification method and apparatus. The method includes: obtaining an input feature map of a to-be-processed image; performing convolution processing on the input feature map based on M convolution kernels of a neural network, to obtain a candidate output feature map of M channels, where M is a positive integer; performing matrix transformation on the M channels of the candidate output feature map based on N matrices, to obtain an output feature map of N channels, where a quantity of channels of each of the N matrices is less than M, N is greater than M, and N is a positive integer; and classify the to-be-processed image based on the output feature map, to obtain a classification result of the to-be-processed image.

Abstract: A connector includes a pair of outer conductors including a first outer conductor and a second outer conductor assembled together and slidable with respect to each other, a pair of center conductors arranged within the outer conductors, an insulation seat, and an elastic element. The center conductors include a first center conductor and a second center conductor assembled together and slidable with respect to each other. The first outer conductor and the first center conductor are fixed to the insulation seat. A first end of the elastic element abuts against the first outer conductor or the insulation seat. The first outer conductor and the first center conductor are both in electrical contact with a first electrical component under a pressing force from the elastic element.

Abstract: The present application discloses an image generation method, a neural network compression method, and a related apparatus and device in the field of artificial intelligence. The image generation method includes: inputting a first matrix into an initial image generator to obtain a generated image; inputting the generated image into a preset discriminator to obtain a determining result, where the preset discriminator is obtained through training based on a real image and a category corresponding to the real image; updating the initial image generator based on the determining result to obtain a target image generator; and further inputting a second matrix into the target image generator to obtain a sample image. Further, a neural network compression method is disclosed, to compress the preset discriminator based on the sample image obtained by using the foregoing image generation method.

Abstract: The present application discloses a neural network search method in the field of artificial intelligence, and the neural network search method includes: obtaining a feature tensor of each of a plurality of neural networks, where the feature tensor of each neural network is used to represent a computing capability of the neural network; inputting the feature tensor of each of the plurality of neural networks into an accuracy prediction model for calculation, to obtain accuracy of each neural network, where the accuracy prediction model is obtained through training based on a ranking-based loss function; and determining a neural network corresponding to the maximum accuracy as a target neural network. Embodiments of the present invention help improve accuracy of a network structure found through search.

Abstract: A cable fixing apparatus adapted to be fixed to an end of a cable includes a first sleeve having a first thread portion at a first end of the first sleeve, a second sleeve having a bottom portion with a through hole at first end of the second sleeve, and a clamp mechanism disposed in the second sleeve between the first thread portion of the first sleeve and the bottom portion of the second sleeve. The cable is inserted into the second sleeve and the first sleeve through the through hole in an insertion direction. The second sleeve is threaded onto the first thread portion at a second end of the second sleeve. The clamp mechanism clamps the cable in response to an engagement of the second end of the second sleeve to the first thread portion to prevent the cable from moving in the first sleeve and the second sleeve.

Abstract: An adapter for mating with a receptacle comprises a cylindrical outer conductor, a columnar center conductor disposed in the outer conductor, and an insulator disposed between the outer conductor and the center conductor. The outer conductor has a recess in an inner wall at a first end of the outer conductor. The insulator electrically isolates the outer conductor from the center conductor.

Abstract: An electrical connector includes a conductive housing having a first receiving portion with an open end, a movable terminal having a first end inserted into the first receiving portion via the open end, the movable terminal having a second receiving portion, and a first elastic piece located in a space defined by the first receiving portion and the second receiving portion. A second end of the movable terminal opposite to the first end protrudes movably out of the first receiving portion against an elasticity of the first elastic piece. The first end of the movable terminal has a first elastic arm elastically abutting an inner side wall of the first receiving portion. The movable terminal is electrically connected to the conductive housing by each of the first elastic piece and the first elastic arm.

Abstract: A connector comprises an adapter and a first receptacle adapted to be assembled onto a first end of the adapter. The adapter includes an outer contact, an inner contact, and an insulator between the outer contact and the inner contact. The first receptacle includes a first outer terminal in electrical contact with a first end of the outer contact, a first inner terminal in electrical contact with a first end of the inner contact, and a first insulative body between the first outer terminal and the first inner terminal. The first outer terminal includes a plurality of first elastic contact structures. Each of the first elastic contact structures has an elastic arm and a first contact point formed on the elastic arm. The first contact point is adapted to elastically electrically contact an inner wall of the first end of the outer contact.

Abstract: A connector adapter includes a first sub-adapter directly and electrically connected with a first end of a connector and a second sub-adapter directly and electrically connected with a second end of the connector. The first end of the connector is electrically connected to a first cable through the first sub-adapter and to a test equipment through the first cable. The second end of the connector is electrically connected to a second cable through the second sub-adapter and to the test equipment through the second cable.

Abstract: A connector comprises a plurality of outer conductors including a first outer conductor and a second outer conductor slidably assembled together, a plurality of center conductors disposed in the outer conductors and including a first center conductor and a second center conductor slidably assembled together, an insulation seat molded on the first outer conductor, and a first elastic element disposed outside of a portion of the outer conductors. The first outer conductor has an electrical contact pin configured to electrically contact a first electronic component. The electrical contact pin is exposed outside from a surface of the insulation seat. An end of the first elastic element abuts the insulation seat and is adapted to apply an axial thrust to the first outer conductor.

Abstract: This application discloses a convolutional neural network-based image processing method and image processing apparatus in the artificial intelligence field. The method may include: receiving an input image; preprocessing the input image to obtain preprocessed image information; and performing convolution on the image information using a convolutional neural network, and outputting a convolution result. In embodiments of this application, the image processing apparatus may store primary convolution kernels of convolution layers, and before performing convolution using the convolution layers, generate secondary convolution kernels using the primary convolution kernels of the convolution layers.

Abstract: A connector comprises a plurality of outer contacts including a first outer contact and a second outer contact that are slidably assembled together, a plurality of inner contacts disposed within the outer contacts, and a first elastic element disposed between the first outer contact and the second outer contact and adapted to exert a first axial pushing force onto the first outer contact. The second outer contact has an outer cylinder and an inner cylinder connected to the outer cylinder. A receiving groove having an annular cross section is defined between the outer cylinder and the inner cylinder. The first outer contact has an elastic latch adapted to be inserted into the receiving groove and adapted to be latched onto an inner wall of the outer cylinder.

Abstract: This application discloses a convolutional neural network-based image processing method and image processing apparatus in the artificial intelligence field. The method may include: receiving an input image; preprocessing the input image to obtain preprocessed image information; and performing convolution on the image information using a convolutional neural network, and outputting a convolution result. In embodiments of this application, the image processing apparatus may store primary convolution kernels of convolution layers, and before performing convolution using the convolution layers, generate secondary convolution kernels using the primary convolution kernels of the convolution layers.

Abstract: A connector includes a pair of outer conductors including a first outer conductor and a second outer conductor assembled together and slidable with respect to each other, a pair of center conductors arranged within the outer conductors, an insulation seat, and an elastic element. The center conductors include a first center conductor and a second center conductor assembled together and slidable with respect to each other. The first outer conductor and the first center conductor are fixed to the insulation seat. A first end of the elastic element abuts against the first outer conductor or the insulation seat. The first outer conductor and the first center conductor are both in electrical contact with a first electrical component under a pressing force from the elastic element.