Abstract: Embodiments of this specification provide methods, apparatuses systems, and computer-readable media for data privacy-preserving training of a service prediction model. In an example training process, a member device performs prediction by using the service prediction model and object feature data held by the member device, and determines, by using a prediction result, update parameters used to update model parameters, where the update parameters include sub-parameters for computational layers of the service prediction model; and divides the computational layers into first-type and second-type computational layers using the sub-parameters; and performs privacy processing on sub-parameters of the first-type computational layers, and outputs processed sub-parameters. Processed sub-parameters of member devices can be aggregated into aggregated sub-parameters.

Abstract: The present disclosure provides a twisting machine for automatic donning, including an outer yarn branch rail arranged at the middle position of the twisting machine, wherein the outer yarn branch rail is positioned above the twisting machine, so as to successively drive the outer yarn carrier assembly to deliver the outer yarn winding package by suspending; further including an inner yarn branch rail arranged at the both sides of the twisting machine, the inner yarn branch rail is positioned above the twisting machine, so as to successively drive the inner yarn carrier assembly to deliver the inner yarn winding package by suspending. The forementioned structure enables the outer yarn winding package and the inner yarn winding package to be delivered high-efficiently in large scale by suspending and conveying, so as to largely reduce the labor intensity of operators and greatly improve production efficiency.

Abstract: The present disclosure provides a whole-process automatic loading-and-unloading and logistics system, including an outer yarn preparation area, an inner yarn preparation area, a weaving area and an in-air conveying rail; the outer yarn preparation area is configured to have a carrier assembly, which delivers an outer yarn winding package to an outer yarn branch rail of a twisting machine in a twisting area via the out yarn main rail of the in-air conveying rail; in the inner yarn preparation area, the carrier assembly delivers an inner yarn winding package to an inner yarn branch rail of the twisting machine in the twisting area via the inner yarn main rail of the in-air conveying rail; in the twisting area, the carrier assembly delivers a yarn product to a yarn product donning rail in the weaving area via the yarn product main rail of the in-air conveying rail.

Abstract: A lifting system for inner yarn rails relates to the technical field of twisting and weaving equipment in factories, the system comprises an in-air conveying rail, and a carrier unit, wherein the in-air conveying rail is configured to convey the carrier unit, and the carrier unit is configured to carry a winding package; an inner yarn-carrier-lifting section is arranged on the in-air conveying rail, the inner yarn-carrier-lifting section is positioned on both sides of a twisting machine; when the inner yarn-carrier-lifting section rises, it communicates with the in-air conveying rail; when the inner yarn-carrier-lifting section descends, it approaches the position of a spindle bucket of the twisting machine. The winding packages delivered each time by one carrier unit can correspond to every spindle position of the twisting machine, dramatically improving the delivery efficiency of the winding packages.

Abstract: A gas-exhausting apparatus includes a gas-exhausting body and a gas-exhausting mechanism. The gas-exhausting mechanism includes a connecting assembly and a gas-ventilating member. The connecting assembly includes first and second connecting members. The first connecting member is provided with a first through hole for communicating with an inside of the battery cell. The second connecting member is provided with a second through hole for communicating with an outside of the battery cell. The first connecting member is configured to attach to a surface of the gas-ventilating member facing the inside of the battery cell. The second connecting member is configured to attach to a surface of the gas-ventilating member facing away from the inside of the battery cell. An orthographic projection of the first through hole on the gas-ventilating member and an orthographic projection of the second through hole on the gas-ventilating member are arranged to be staggered from each other.

Abstract: Disclosed in the present disclosure is a double-deck multi-span bridge construction method. According to the double-deck bridge construction method of the present disclosure, construction is carried out by using a method of disassembling a support jig frame in a graded and span-separated mode, an upper chord jig frame and a lower chord jig frame can be used in a recycle manner, and construction costs are reduced. In addition, a construction period of building the support jig frame is shortened, and other construction operations can be synchronously carried out on a span in which the jig frame is disassembled, for example, fire retardant coating construction can be carried out on a mounted bridge deck after the jig frame is disassembled, and the construction period of a double-deck multi-span bridge is effectively shortened.

Abstract: For the problem of difficult roof anchor withdrawal at the end of a working face, a static crushing directional anchor withdrawal method at the end of a large-mining-height working face is provided. The method includes the following steps: designing key parameters of boreholes for static crushing, including a borehole position, a spacing between boreholes, a borehole depth, and a borehole diameter at the end of a working face; then determining parameters of a static crushing agent grouting technology device for the construction of a static crushing process; and carrying out an anchor withdrawal operation at the end.

Abstract: A display module, including: a display panel; and a color film layer and a micro-lens layer, provided on a light-emitting side of the display panel; where, the color film layer includes a plurality of filter portions, and the micro-lens layer includes a plurality of first converging lenses and a plurality of second converging lenses; a gap is provided between two adjacent first converging lenses, and an orthographic projection of a first converging lens on the display panel is located within an orthographic projection of a filter portion on the display panel; a second converging lens is provided in the gap between two adjacent first converging lenses, and a vertex of the second converging lens is located between a vertex of the first converging lens and a bottom surface of the first converging lens.

Abstract: The present application relates to an all-optical detector and detection system, a response time test system, and a manufacturing method. The all-optical detector comprises a micro-nanofiber and an optical resonant cavity. The micro-nanofiber comprises transition regions and a uniform region. The uniform region is connected to the transition regions. The optical resonant cavity is provided in the uniform region. The optical resonant cavity is made of a semiconductor material. The all-optical detector provided in the present application detects light by means of the change of a resonance peak, achieves all-optical detection, and has a high signal-to-noise ratio.

Abstract: A predictive contrastive representation method for multivariate time-series data processing includes: mapping temporal coding information at a current moment and future situational information by using a logarithmic bilinear model to obtain a similarity; training the similarity according to a noise contrastive estimation method and prediction situational label data, and constructing, based on a training result, a predictive contrastive loss function of the temporal coding information at the current moment and the future situational information; sampling the prediction situational label data based on a corresponding optimal loss in the predictive contrastive loss function, optimizing the predictive contrastive loss function by using a direct proportion property between the sampling probability and the similarity, constructing mutual information between the temporal coding information at the current moment and the future situational information based on the optimized predictive contrastive loss function, and pe