Abstract: Provided are a method and device for data processing and a storage medium. The method includes: inputting labeled training data in a training set into a preset model to be trained and updating the model parameters of the preset model; inputting labeled verification data in a verification set into the preset model after the model parameters are updated to obtain a first prediction label; obtaining a verification loss value based on a difference between the first prediction label and a marked label of the labeled verification data; determining an auxiliary loss value based on current structural parameters of the preset model; determining whether to stop training the preset model based on the verification loss value and the auxiliary loss value; and classifying data to be classified based on a target network model constructed by a network structure included in the trained preset model to obtain a classification result.

Abstract: A display substrate and a display device. In the display substrate, at least one of the inter-opening region, the first opening peripheral region and the second opening peripheral region includes a first virtual sub-pixel; the first signal line extends along a first direction and includes a first portion passing through the first opening peripheral region, the inter-opening region and the second opening peripheral region; the first portion passes through the first virtual sub-pixel, and the first virtual sub-pixel includes a first compensation capacitor, a first plate of the first compensation capacitor is in a same layer as the first portion of the first signal line and electrically connected with the first portion of the first signal line, and in a same layer as the second plate of the storage capacitor; the second plate is in a different layer from, insulated from, and overlaps with the first plate.

Abstract: In one example, a semiconductor device includes a conductive layer, composite structures, conductive posts and first pads. The composite structures may be located on the conductive layer and stacked in a direction perpendicular to the plane in which the conductive layer is located. The composite structure may include a chip, an insulating layer surrounding around the chip, and at least one second pad electrically connected with the chip. The second pad is located on the insulating layer. The second pads of the composite structures are at different locations in the first direction. The first direction is perpendicular to the thickness direction of the composite structures. The conductive posts are located in the insulating layer of the composite structures and each conductive post is connected with one of the second pads and one of the first pads.

Abstract: The present utility model proposes a refrigeration display cabinet, comprising: a cabinet; a display rack fixed inside the cabinet; an air channel arranged at least partially surrounding the display rack; a partition mounted in the air channel to separate the air channel into an inner air channel and an outer air channel, wherein the partition is provided with an opening for connecting the inner air channel and the outer air channel; an evaporator arranged in the inner air channel, wherein the side of the evaporator is located at or near the opening; and a damper assembly fixed at the opening of the partition, wherein the damper assembly is capable of opening or closing the opening. The refrigeration display cabinet according to the present utility model can defrost the evaporator simultaneously from the bottom and side thereof during the defrosting mode, thus improving defrosting efficiency.

Abstract: A process system of a follower fixture for tank moving support is provided and includes a conveying table slidably provided with a bearing platform through a screw slide rail assembly and provided with an outer frame assembly; a lifting platform is slidably installed in the outer frame assembly; positions of the lifting platform and the bearing platform are corresponded up and down; the outer frame assembly is slidably provided with a robot component; both sides of the conveying table are provided with loading platforms; workpieces are placed on one loading platform, and a fixture system is placed on the other loading platform; the fixture system and the workpieces are both placed in the moving range of the robot component, and positions of the fixture system and the workpieces are corresponded. problems including processing flexibility cannot be realized can be solved.

Abstract: Fuse programming circuits, devices and methods. In some embodiments, a fuse circuit can include a fuse pad configured to receive a voltage, a fuse having a first end coupled to the fuse pad and a second end coupled to a switching element configured to enable a current to pass from the fuse pad to a ground potential.

Abstract: A switching component and switch assembly. The switching component comprises a first control node, a common node, a plurality of intermediate nodes, a second control node, and a capacitive node; a plurality of transistors connected in series between the control node and the common node, one of the plurality of intermediate nodes being defined between each series connected pair of transistors, each transistor of the plurality of transistors having a gate coupled to the second control node; and a plurality of capacitive components, one capacitive component being coupled between each intermediate node and the capacitive node, a voltage at the capacitive node being configured to be varied with a voltage at the second control node such that, at each intermediate node, the capacitive component is configured to accrue an opposite charge to the transistors.

Abstract: Methods, systems, and devices for read performance techniques for time retention are described. A memory system may store data in a block of memory cells and perform a power cycle operation. Based on performing the power cycle operation, the memory system may determine a first voltage offset associated with the block of memory cells by executing a first read command using an auto-read calibration operation. Based on the first voltage offset, and, in some examples, one or more additional voltage offsets, the memory system may calculate a retention time of data stored in the block of memory cells. The memory system may adjust a read voltage based on the retention time and perform one or more additional read commands.

Abstract: A preparation method of a rare earth permanent magnetic material comprises one or more times of cryogenic treatment and tempering treatment. When the preparation method comprises one time of cryogenic treatment and tempering treatment, the cryogenic treatment is performed between sintering treatment and the tempering treatment, or is performed after the tempering treatment. When the preparation method comprises two or more times of cryogenic treatment and tempering treatment, at least one time of cryogenic treatment is performed between sintering treatment and the tempering treatment, and at least one time of cryogenic treatment is performed after the tempering treatment. The preparation method can refine and homogenize the structure and realize quick and convenient treatment, and can effectively improve the mechanical performance of the rare earth permanent magnetic material and maintain good magnetic performance.

Abstract: Circuits, methods, and devices for protecting against accidental fuse programming are discussed herein. For example, a fuse circuit may include a fuse, a first switch coupled to a first point and coupled in series with the fuse, and a second switch coupled in series with the fuse between the fuse and a second point.

Abstract: A polishing machine with an automatic feed and discharge function includes a machine frame, a clamping mechanism, a polishing shaft, a feed mechanism and a discharge mechanism. The clamping mechanism, the discharge mechanism, the polishing shaft and the feed mechanism are mounted on the machine frame. Multiple polishing heads are mounted on the polishing shaft. The feed mechanism includes a feed frame, a sliding platform, a sliding drive member, multiple columns of storage mechanisms for storing polishing pads, and multiple columns of push mechanisms, and the number of the columns of the storage mechanisms matches the number of the columns of the push mechanisms. The feed frame is mounted on the machine frame. The sliding platform is slidably arranged on the feed frame. The storage mechanisms are mounted on the sliding platform. The push mechanisms are mounted on the feed frame and are located above the storage mechanisms.

Abstract: Provided display panel includes a first substrate and a light-emitting element array. The first substrate includes a supporting substrate, a light-shielding layer disposed on the supporting substrate and provided with multiple light-shielding openings, and a eutectic layer disposed on the supporting substrate and provided with multiple eutectic material blocks corresponding to the multiple light-shielding openings respectively. In the direction perpendicular to the first substrate, a eutectic material block overlaps a corresponding light-shielding opening. The light-emitting element array includes multiple light-emitting elements, and is electrically connected to the first substrate.

Abstract: Provided are a light-emitting diode module, a preparation method therefor, a transfer method for a light-emitting element, a display panel, and a display device. The light-emitting diode module includes a first substrate, a sacrificial protective layer, and multiple light-emitting elements. The first substrate has multiple first regions arranged at intervals and second regions, where a second region is located between adjacent first regions. The sacrificial protective layer is located on a side of the first substrate and at least partially located in the second region. The multiple light-emitting elements are disposed in the multiple first regions and located on the same side of the first substrate as the sacrificial protective layer.

Abstract: Methods, systems, and devices for memory block utilization in memory systems are described. A system configured to allow a memory device to group or segment a memory block into two or more sub-memory blocks, which can be independently programmed is described herein. For example, a host system may determine a configuration of a memory array, and communicate the configuration information to the memory system, and transmit a command for an operation to the memory system. In some examples, the memory system may utilize the memory array configuration information and determine to segment the blocks of memory cells into sub-blocks. By segmenting the memory block into sub-blocks, the memory device may maintain its memory block density while supporting efficient programming of blocks of the memory array.

Abstract: Systems and methods for dual-domain self-supervised learning for accelerated non-Cartesian magnetic resonance imaging reconstruction are provided. The present techniques provide a method for training a machine-learning model that receives magnetic resonance (MR) data and generates a reconstruction of the MR data. The machine-learning model can be trained based on a set of losses comprising a first loss value corresponding to a frequency-domain and a second loss value corresponding to an image-based domain. The training process can be a self-supervised training process that can utilize under-sampled and non-Cartesian MR data. The machine-learning model is trained by optimizing both data consistency in the frequency domain and appearance consistency in the image-based domain.

Abstract: Methods and systems for image processing are provided. The method may include obtaining image data generated by an image acquisition device; generating a preliminary image by processing the image data; and generating a target image by using an image processing model to process the preliminary image according to an optimization algorithm. The image processing model includes a first sub-model and a second sub-model. The first sub-model is configured to determine a first optimization term related to a likelihood term of an objective function, and the second sub-model is configured to determine a second optimization term related to a regularization term of the objective function, wherein the second sub-model is a trained machine-learning model.

Abstract: An attachment for a stand mixer includes an ice block-receiving chamber having an inner cavity and a lower side defining an output opening and a blade in a fixed position along a portion of the opening and partially extending into the inner cavity. The attachment for a stand mixer further includes a plunger mounted with respect to the ice block-receiving chamber so as to be moveable into the inner cavity toward the lower side along an axis and to be rotatable around the axis, the blade having an edge that extends axially away from the axis and a drive shaft having an input end coupleable to a power output of the stand mixer and a drive end mechanically coupled with the plunger to drive rotation thereof about the axis.

Abstract: Devices and methods for operating a charge pump. In some implementations, a charge pump module includes a clock circuit configured generate to a first clock signal and a second clock signal, the first clock signal having a lower frequency than the second clock signal. The charge pump module also includes a driving circuit configured to generate a first set of clock signals based on the first clock signal and a second set of clock signals based on the second clock signal, the driving circuit coupled to the clock circuit. The charge pump module further includes a charge pump core including a set of capacitances, the charge pump core configured to charge the set of capacitances based on the first set of clock signals and the second set of clock signals.

Abstract: A display panel, a method for manufacturing the display panel, a display screen, and a display device are provided. The display panel includes a substrate and multiple first wires. The substrate includes a first surface and a second surface that are opposite. The first surface faces a display face of the display panel. The multiple first wires are arranged on a side of the first surface away from the second surface. The second surface includes a first division and a second division. A maximum distance between adjacent first wires in the first division is H1, and a maximum distance between adjacent first wires in the second division is H2. H1 is greater than H2. The distance between adjacent first wires in the first division is increased to be greater than the distance between adjacent first wires in the second division.