Abstract: A monitoring system based on a digital converter station includes a first monitoring terminal deployed at a first-level monitoring side, a second monitoring terminal deployed at a second-level monitoring side, and a third monitoring terminal deployed at a third-level monitoring side; the monitoring terminal at each level includes a communication module, a human-machine interaction module, a device status monitoring module, a device alarm management module, a video fusion processing module, and a data transmission adjustment and control module.

Abstract: This application provides a method for training a neural network model and an apparatus. The method includes: obtaining annotation data that is of a service and that is generated by a terminal device in a specified period; training a second neural network model by using the annotation data that is of the service and that is generated in the specified period, to obtain a trained second neural network model; and updating a first neural network model based on the trained second neural network model. In the method, training is performed based on the annotation data generated by the terminal device, so that in an updated first neural network model compared with a universal model, an inference result has a higher confidence level, and a personalized requirement of a user can be better met.

Abstract: Described herein are methods and compositions for the treatment of conditions, diseases, or disorders associated with thyroid hormone receptor activity. The methods and compositions disclosed herein include the use of at least one thyroid hormone receptor agonist.

Abstract: The present disclosure relates to the technical field of multimedia, and in particular, to a control method and apparatus for multimedia content display, an electronic device, and a medium. First multimedia content is displayed on a first page; and in response to a terminal device changing in a preset attitude, the first page is switched to a second page, and second multimedia content is displayed on the second page. That is, the switching of the first page to the second page is implemented by triggering the attitude change of the terminal device, thereby improving the interestingness of user interaction.

Abstract: Embodiments relate to decoding communications with token sky maps. At least one electronic communication including emoticons having a non-original meaning is received. A candidate meaning is determined for the emoticons having the non-original meaning in the at least one electronic communication based at least in part on token neighborhood distribution structures. The candidate meaning for the emoticons having the non-original meaning is caused to be displayed on at least one device.

Abstract: A method, computer system, and a computer program product for syllable-based pronunciation help are provided. An input text in a first language may be received. A selection of a target language that is different from the first language may be received. From the target language, syllables with a pronunciation most closely matching a pronunciation of the input text in the first language are obtained. The obtaining is based on a comparison of one or more spectrograms for the input text with one or more spectrograms for text of the target language. The obtained syllables in the target language are presented.

Abstract: A method of making photolithography mask plate is provided. The method includes: providing a carbon nanotube composite structure, wherein the carbon nanotube composite structure comprises a carbon nanotube layer and a chrome layer coated on the carbon nanotube layer; locating the carbon nanotube composite structure on a substrate to expose partial surfaces of the substrate; and depositing a cover layer on the carbon nanotube composite structure.

Abstract: A method for making a field effect transistor includes providing a graphene nanoribbon composite structure. The graphene nanoribbon composite structure includes a substrate and a plurality of graphene nanoribbons spaced apart from each other. The substrate includes a plurality of protrusions spaced apart from each other, and one of the plurality of graphene nanoribbons is on the substrate and between two adjacent protrusions. An interdigital electrode is placed on the graphene nanoribbon composite structure, and the interdigital electrode covers the plurality of protrusions and is electrically connected to the plurality of graphene nanoribbons.

Abstract: A method of making photolithography mask plate is provided. The method includes: providing a carbon nanotube layer on a substrate; depositing a chrome layer on the carbon nanotube layer, wherein the chrome layer includes a first patterned chrome layer and a second patterned chrome layer, the first patterned chrome layer is located on the carbon nanotube layer, and the second patterned chrome layer is deposited on the substrate corresponding to holes of the carbon nanotube layer; transferring the carbon nanotube layer with the first patterned chrome layer thereon from the substrate to a base, and the carbon nanotube layer being in contact with the base; and depositing a cover layer on the first patterned chrome layer.

Abstract: A method for controlling a device for automatically adjusting an airway opening body position is provided. The device includes a horizontal base plate, a head support block, a back support plate, a neck support apparatus, a head cover assembly, and a programmable logic controller (PLC). The neck support apparatus is positioned between the head support block and the back support plate. The PLC is configured to controls a stroke of an electric cylinder according to the following equations: ?=1.235?+?, and ?=KX+B+C, where ? is a body position angle, the body position angle is an angle between a positive projection line of a connecting line from a mandibular angle to an external acoustic meatus on a symmetrical surface of a human body and the back support plate, and ? is a preset value ranging from 90° to 100°.

Abstract: A first device is logged in with a first user account. The first device displays an avatar of a second user account in a first user interface of the first device. In response to an interaction instruction triggered on the avatar of the second user account, the first device generates an interaction message according to a first field corresponding to the first user account, an action description field corresponding to the interaction instruction, and a second field corresponding to the second user account. The first device transmits the interaction message to a second device that is logged in with the second user account.

Abstract: The present invention provides a vertical feeding and wafer inserting integrated machine, comprising a rack and a silicon wafer transfer device provided on the rack; a carrying-transporting water tank for carrying and transporting silicon wafers is provided on one end of the rack; the rack is further provided with a feeding unit for conveying the silicon wafers in the carrying-transporting water tank to a silicon wafer transfer belt a wafer inserting unit for inserting the silicon wafers into wafer baskets is provided on one end, away from the feeding unit, of the silicon wafer transfer device, and a wafer basket arraying mechanism is provided on the other end of the wafer inserting unit; and a washing manipulator for transferring grouped wafer baskets to a washing procedure is further correspondingly provided on the rack.

Abstract: System and methods for compressing image-to-image models. Generative Adversarial Networks (GANs) have achieved success in generating high-fidelity images. An image compression system and method adds a novel variant to class-dependent parameters (CLADE), referred to as CLADE-Avg, which recovers the image quality without introducing extra computational cost. An extra layer of average smoothing is performed between the parameter and normalization layers. Compared to CLADE, this image compression system and method smooths abrupt boundaries, and introduces more possible values for the scaling and shift. In addition, the kernel size for the average smoothing can be selected as a hyperparameter, such as a 3×3 kernel size. This method does not introduce extra multiplications but only addition, and thus does not introduce much computational overhead, as the division can be absorbed into the parameters after training.

Abstract: System and methods for compressing image-to-image models. Generative Adversarial Networks (GANs) have achieved success in generating high-fidelity images. An image compression system and method adds a novel variant to class-dependent parameters (CLADE), referred to as CLADE-Avg, which recovers the image quality without introducing extra computational cost. An extra layer of average smoothing is performed between the parameter and normalization layers. Compared to CLADE, this image compression system and method smooths abrupt boundaries, and introduces more possible values for the scaling and shift. In addition, the kernel size for the average smoothing can be selected as a hyperparameter, such as a 3×3 kernel size. This method does not introduce extra multiplications but only addition, and thus does not introduce much computational overhead, as the division can be absorbed into the parameters after training.

Abstract: In a job management environment comprising a plurality of job systems and a scheduler for scheduling a job submitted to the job management environment to a job system for running, a processor, in a first job system, intercepts, from outside of a first container in the first job system, a first job from being sent to the scheduler. A processor, in the first job system, determines whether the first job is submitted from a container in the first job system. In response to a first determination that the first job is submitted from a container in the first job system, a processor, in the first job system, determines contexts of the first job, the contexts of the first job including a context related to the first container. A processor, in the first job system, sends the first job together with the contexts of the first job to the scheduler.

Abstract: Techniques for training a neural network having a plurality of computational layers with associated weights and activations for computational layers in fixed-point formats include determining an optimal fractional length for weights and activations for the computational layers; training a learned clipping-level with fixed-point quantization using a PACT process for the computational layers; and quantizing on effective weights that fuses a weight of a convolution layer with a weight and running variance from a batch normalization layer. A fractional length for weights of the computational layers is determined from current values of weights using the determined optimal fractional length for the weights of the computational layers. A fixed-point activation between adjacent computational layers is related using PACT quantization of the clipping-level and an activation fractional length from a node in a following computational layer.

Abstract: System and methods for compressing image-to-image models. Generative Adversarial Networks (GANs) have achieved success in generating high-fidelity images. An image compression system and method adds a novel variant to class-dependent parameters (CLADE), referred to as CLADE-Avg, which recovers the image quality without introducing extra computational cost. An extra layer of average smoothing is performed between the parameter and normalization layers. Compared to CLADE, this image compression system and method smooths abrupt boundaries, and introduces more possible values for the scaling and shift. In addition, the kernel size for the average smoothing can be selected as a hyperparameter, such as a 3×3 kernel size. This method does not introduce extra multiplications but only addition, and thus does not introduce much computational overhead, as the division can be absorbed into the parameters after training.

Abstract: A method and system determines in real time the bearing capacity of a foundation tamped by a high-speed hydraulic tamper. Four wireless acceleration sensors are arranged uniformly along tamping plate edges, sensor position tamping points are determined; the soil is tamped, the plate peak acceleration slows, tends to, and reaches stabilization in a range, a relationship curve between the tamping number and plate peak acceleration is determined; different loads are applied to the foundation to obtain corresponding settlements, coordinate axes are established, points are drawn according to each test data group and sequentially connected with a smooth curve to obtain a settlement-load curve, and the curve is fitted; a tamping number and foundation bearing capacity relationship is obtained; the two relationship curves are combined to obtain a relationship curve, and the foundation bearing capacity magnitude at a certain moment during the tamping operation is determined by using the acceleration index.

Abstract: Systems and methods herein describe an image compression system. The image compression system generates a first generative adversarial network (GAN), identifies a threshold, based on the threshold, generates a second GAN by pruning channels of the first GAN, trains the second GAN using similarity-based knowledge distillation from the first GAN, and stores the trained second GAN.

Abstract: A method and system determines in real time the bearing capacity of a foundation tamped by a high-speed hydraulic tamper. Four wireless acceleration sensors are arranged uniformly along tamping plate edges, sensor position tamping points are determined; the soil is tamped, the plate peak acceleration slows, tends to, and reaches stabilization in a range, a relationship curve between the tamping number and plate peak acceleration is determined; different loads are applied to the foundation to obtain corresponding settlements, coordinate axes are established, points are drawn according to each test data group and sequentially connected with a smooth curve to obtain a settlement-load curve, and the curve is fitted; a tamping number and foundation bearing capacity relationship is obtained; the two relationship curves are combined to obtain a relationship curve, and the foundation bearing capacity magnitude at a certain moment during the tamping operation is determined by using the acceleration index.