25results about How to "Improve generation effect" patented technology

The application discloses a SiC reactor for cyclohexanone aminooxidation and a design method thereof, and relates to the technical field of chemical reactors. The SiC reactor comprises a reactor shell, a plurality of reaction channels, a plurality of composite plates, and a plurality of reaction channels. The axis direction of the reaction channel is parallel to the axis direction of the reactor shell. The plurality of composite plates are arranged in the plurality of reaction channels. The flat plate and the corrugated plate made of silicon carbide are arranged in the axial direction of the reaction channel and matched with the rhombic through-hole structure. The fluid in the reaction channel can be disturbed, the radial diffusion effect of the fluid is improved, and the effective residence time of the fluid is prolonged. Based on the excellent heat conduction performance of silicon carbide and the plate combination structure, the heat exchange between the wall surface and the fluid is strengthened, the temperature field distribution in the channel is optimized, the mass transfer and heat transfer conditions of the material are improved, and the generation effect of cyclohexanone oxime is improved.

The present disclosure provides a script generation method and device, computer equipment and a storage medium, wherein the method comprises: obtaining a plurality of to-be-screened scripts; inputting the plurality of to-be-screened scripts into a pre-trained script generation model to obtain target scripts generated by the script generation model based on a plurality of to-be-screened scripts corresponding to a same target theme; wherein the script generation model is trained based on target sample data, any target sample data is constructed based on a first sample script and a confusion sample script corresponding to the first sample script, the confusion sample script comprises a second sample script of the same theme type as the target theme and a third sample script of a different theme type from the target theme.

This application relates to the field of multimodal technology, and discloses a method, apparatus, device, and storage medium for point cloud generation based on multimodal information. The method includes: generating a first loss of a point cloud generator based on a first predicted point cloud set, a first real point cloud set, and a first loss model; generating a second loss of the point cloud generator based on a second predicted point cloud set, a second real point cloud set, and a second loss model; generating a third loss of the point cloud generator based on a third predicted point cloud set, a third real point cloud set, and a third loss model; generating a total loss based on the first loss, second loss, third loss, and a total loss function; and generating a complete point cloud of the object to be completed based on a cross-fusion module, a seed generator, and an updated point cloud generator when the total loss meets preset conditions. This application can take into account the overall outline, main structure, and local details of the object to be completed, which is beneficial to improving the generation effect of the complete point cloud of the object to be completed.

This invention discloses a code generation method and system for a specific programming language in a low-resource domain, belonging to the field of code generation technology. The method includes: generating an intermediate representation and domain feature description of historical code, and constructing a vector library based on the historical code; retrieving the vector library based on the domain feature description of the target task to obtain several historical tasks; and generating code for the target task based on the domain feature description of the target task and the retrieved historical tasks. This invention can accurately generate code for a target domain using a low-resource domain.

The invention discloses a carrier supported multi-metal sulfide state nickel-based catalyst as well as a preparation method and application thereof. The catalyst is composed of a porous carrier loaded NixMySz active phase, M is a combination of at least one metal element of Co, Fe, Cu, Zn or Mn, multiple metal components form a synergistic structure in the vulcanization process, and through construction of Ni-S-M bonds, pinning effect generation and electron interaction, controllable adjustment of the sulfide structure is achieved, and the stability of the catalyst is improved. The catalyst is obtained by loading a nickel-second metal precursor on a porous carrier, roasting to form an oxide, vulcanizing in a sulfur-containing atmosphere, and activating. The obtained catalyst has an adjustable Ni: M molar ratio, a sheet-like or petal-like nanostructure and high-density sulfur defects. The catalyst is suitable for decarbonylation-dehydration reaction of bio-based fatty acid or derivatives thereof, and can efficiently convert C10-C20 fatty acid, methyl ester or ethyl ester thereof, and animal and vegetable oil or waste oil containing the components into long-chain olefin at 200-350 DEG C under 0.1-3 MPa.

The application provides a display picture generation method and system based on a smart sound box, which is suitable for the field of image processing technology. The method comprises the following steps: generating initial display picture information according to user interaction instruction information, interaction scene image information, a user interaction instruction analysis model and an interaction scene image matching model; obtaining target display picture center coordinate information according to the user interaction instruction information, the initial display picture information and historical display picture information; obtaining target display picture scaling multiple information according to the target display picture center coordinate information, display picture size information and display picture reference point coordinate information; and generating target display picture information according to the initial display picture information, the target display picture center coordinate information, the target display picture scaling multiple information and a display picture generation model. The application improves the pertinence and adaptability of the display picture generation of the smart sound box, thereby optimizing the visual experience of the user in the interaction process with the smart sound box.

The invention relates to the field of intelligent medical treatment, in particular to a children disease cause analysis method and device based on interlayer information reuse and a program product. The method includes: acquiring a child case data set and a disease tag; inputting the child case data set and the disease label to a to-be-trained neural network model for training to obtain a child pathogenesis analysis model; the neural network model carries out parameter sharing through interlayer similarity calculation to reduce the total parameter quantity of the model, the model comprises a first attention layer and a second attention layer, the similarity of the first attention layer and the second attention layer is calculated, and when the similarity of the first attention layer and the second attention layer is larger than a preset threshold value, the first attention layer and the second attention layer are selected. If the layer is judged to be a functional similar layer, endowing the parameters of the first attention layer to a second attention layer; otherwise, the second attention layer obtains parameters through vector feature learning. According to the model, the understanding ability and the generation ability of the child disease data can be met, and the time and space overhead needs to be reduced.

The invention belongs to the technical field of mechanical health prediction, and particularly discloses a power device health degree prediction method based on deep learning. The method comprises the following steps: dividing signals of detection points according to working condition types; inputting various working condition domain data into a variational auto-encoder to which domain confrontation constraints are applied, and generating feature representations irrelevant to working condition types; carrying out trend evaluation on the time sequence of all feature representations, and screening out a feature representation subset which is most relevant to the health degradation process; deducing the health degree of the detection point according to each feature representation in the feature representation subset; representing the health degree of the power device by the weighted sum of the health degrees of all the detection points; and extracting local features of the health degree of the power device through the time domain convolutional network, and obtaining a predicted value of the health degree of the power device after the long-short-term memory network captures a dynamic evolution rule based on the local features. Compared with the prior art, the method is higher in health prediction and evaluation precision of the power device in a multi-working-condition and strong-interference environment.

This invention relates to a method for preparing double perovskite La2CoFeO6 and its application in degrading odorous pollutants, comprising the following steps: dissolving La(NO3)3·6H2O, Co(NO3)2·6H2O, and Fe(NO3)3·9H2O in a mixed solution of deionized water and ethylene glycol or deionized water and citric acid, adjusting the pH, and then mixing thoroughly; placing them separately in a high-pressure reactor and heating or evaporating the moisture and drying in an oven to obtain precursor powder or dry gel; then calcining at high temperature and cooling to obtain double perovskite La2CoFeO6 powder. The R3c phase La2CoFeO6 material provided in this application, by combining crystal structure regulation with oxygen vacancy defect engineering, establishes a synergistic enhancement mechanism of "structure optimization + defect regulation" for double perovskite materials, significantly improving the photoresponse capability and catalytic performance of the material, providing a new technical path and material design idea for the design of subsequent high-performance photocatalysts for odor pollution control.

Provided are a training sample set generation method, a deep generative model training method and device. In the training sample set generation, an original sample set is divided into a plurality of aggregated sample sets with weak randomness, and a training sample set is constructed based on the aggregated sample sets. In the model training, the training sample set is used as a first training sample set; the deep generative model is iteratively trained based on the training sample set and a preset mode until a preset condition is reached, and in each iteration round: an aggregated training sample set is sampled from the training sample set; a corresponding third sample set is generated by sampling from the deep generative model based on each training sample in the aggregated training sample set; and a maximum mean difference between the aggregated training sample set and the third sample set is used as a loss function for optimization to update parameters of the deep generative model. Thus, in the model training, a method of first division and then matching is adopted for the sample set, and the model generation effect is better.

This application relates to the field of data processing technology, and discloses a graph structure generation method and a memory computing accelerator based on a memory computing accelerator. The graph structure generation method based on the memory computing accelerator includes: a node generation module calculating the hidden state of the current node and sending it to an edge probability generation module; the edge probability generation module outputting a probability vector based on the hidden state and sending it to a Bernoulli sampling module; the Bernoulli sampling module outputting a decision vector based on the probability vector and sending it to an edge storage module; and the edge storage module receiving the decision vectors corresponding to all nodes and generating a target graph structure based on the decision vectors corresponding to all nodes. This invention can reduce computational complexity, reduce data movement, save storage space, and improve energy efficiency, thereby improving the graph structure generation effect.

The application provides a medical image cross-modality generation method based on double-contrast frequency domain decomposition and adaptive diffusion, which comprises the following steps: acquiring MRI samples and preprocessing, constructing a k-space decomposition and enhancement process, constructing a k-space double-flow adaptive degradation network, and constructing a k-space double-flow adaptive prediction network. Through data enhancement technology and an adaptive Gaussian filter set, different frequency components are separated and enhanced, common edge features are highlighted, and the quality of generated information is improved. By using a residual noise diffusion model, high-resolution images with rich details and high diversity are generated while maintaining smooth areas and large-scale structures of images. By introducing anatomical structure consistency constraints and multi-modality fusion optimization strategies, the complementary information of T1 and T2 weighted MRI is fully utilized, and the generation effect and visual authenticity of the medical image cross-modality generation in a complex scene are significantly improved.

The application provides a fusion polypeptide and application thereof, takes MG7-Ag single epitope as a target antigen (MG1) of a digestive tract tumor, and is composed of a T helper epitope peptide and three MG1 in series which are connected by a covalent amide bond at an N terminal, wherein, an adapter lysine residue is removed at a splicing position of adjacent MG1 segments of the MG3. The application covalently couples the T helper epitope peptide at the N terminal of the MG3, and relies on a Th epitope to mediate an MHC II class pathway to present, strongly activates a helper T cell, greatly improves an anti-MG1 specific IgG production capacity, and solves a natural immunogenicity deficiency defect of the MG1 from a molecular structure level.

This invention relates to the field of cell biology, and discloses a culture medium and method for inducing physiological myocardial hypertrophy based on carnosine. The culture medium for inducing physiological myocardial hypertrophy contains carnosine. By introducing an appropriate concentration of carnosine into a cardiomyocyte system, this invention can induce a series of phenotypic changes in cardiomyocytes related to physiological myocardial hypertrophy, including increased cardiomyocyte area, increased mitochondrial number, enhanced oxidative metabolic capacity, and increased ATP production capacity. These changes can improve the energy metabolism level of cardiomyocytes, thus providing a new technical solution for studying myocardial adaptive remodeling.

The application relates to the technical field of image generation, and provides a cross-modal image generation method, device, equipment, product and medium. The method comprises the following steps: standardizing a synthetic aperture radar image to obtain a standardized radar image; extracting prior information related to a scene structure in the standardized radar image to construct a structure prior field; inputting the standardized radar image and the structure prior field into a cross-modal image generation model to obtain an optical image output by the cross-modal image generation model; and the cross-modal image generation model is obtained by sequentially performing multi-step training and single-step training on a prediction velocity field based on historical standardized radar images and historical structure prior fields. In the generation process of the synthetic aperture radar image into the optical image, the application can simultaneously consider the structure keeping capability and the generation efficiency, and effectively improves the generation effect.

The present application relates to a light and dark dual-mode regulation wound microenvironment material and its preparation method and application, and belongs to the technical field of biomedical functional materials. The chemical formula of the material is MoO 3‑x / Ho2O3 / N-C, MoO 3‑x / Ho2O3 nanoparticles are loaded, coated or embedded in a nitrogen-doped carbon skeleton, wherein 0<=x<=2. The present application also provides a preparation method of the material and its application in preparing treatment of infected diabetic wounds, chronic non-healing wounds, bacterial infection wounds, oxidative stress related tissue damage and vascular repair products. The light and dark dual-mode regulation wound microenvironment material provided by the present application has SOD-like and CAT-like nanoscale enzyme activity, can inhibit bacteria and kill bacteria, reduce oxidative damage, and protect fibroblasts, endothelial cells and macrophages. At the same time, it covers multiple pathological links such as bacterial infection, lactic acid accumulation, ROS excess, inflammation persistence and insufficient vascularization in infected diabetic wounds, and has more complete wound microenvironment regulation ability.

The application provides a spatiotemporal sequential regulation photothermal injectable hydrogel and a preparation method and application thereof, and belongs to the technical field of biological medicine. The hydrogel is composed of quaternized methacrylated chitosan, benzaldehyde-terminated polyethylene glycol, a photoinitiator and polydopamine-modified nano-hydroxyapatite, and forms a double-network crosslinked structure through a Schiff base reaction and a blue light-induced radical polymerization. The hydrogel has good injectability, mechanical properties and biocompatibility. Under near-infrared light irradiation, spatiotemporal sequential treatment can be realized by regulating the light power in stages: high-power irradiation produces high temperature to rapidly kill bacteria, and low-power irradiation produces mild photothermal stimulation to promote osteogenesis and angiogenesis. The hydrogel has the functions of antibiosis, osteogenesis promotion and angiogenesis promotion, and can be used for precise repair of infected bone defects, and has a wide clinical application prospect.

The invention relates to a wide-angle oblique incidence-oriented near-to-eye display system and a hologram generation method, and the method comprises the steps: obtaining the initial phase distribution of a reflection-type silicon-based liquid crystal spatial light modulator plane, and enabling the initial phase distribution to have a physical resolution; a virtual oversampling grid is constructed, the initial phase distribution is mapped to an expanded high-resolution grid, and the sampling interval of the expanded high-resolution grid is smaller than the physical sampling interval; constructing a modulation light field on the expanded high-resolution grid, and superposing a linear phase term corresponding to oblique incidence; carrying out propagation calculation on the modulated light field by adopting a Shifted angular spectrum method to obtain high-resolution light field distribution of a target plane; the high-resolution light field distribution of the target plane is mapped back to the physical resolution through average downsampling, a reconstructed image is obtained, and the method has the advantages that on the premise that high quality is kept, the calculation efficiency and the generation acceleration capacity are greatly improved, and the technical effect that hologram generation is rapidly completed is achieved.

The present application relates to the field of image processing, and relates to a kind of clothing model image generation method and system, the method includes: obtaining clothing commodity graph, and different types of features of corresponding clothing are extracted according to clothing commodity graph;Obtain fusion feature;Generate candidate model set, and then obtain candidate model;The body shape parameters of each candidate model are adjusted using 3D human posture model, and the posture parameters of each candidate model are adjusted;The similarity of multiple types of features between each candidate model and clothing commodity graph is calculated respectively, the similarity of each type feature is weighted and summed to obtain the comprehensive similarity corresponding to each candidate model;And the model with the comprehensive similarity greater than similarity threshold is regarded as optimal model.The method of the present application can improve the generation efficiency of clothing model image and reduce the difference between the clothing of clothing model image and the clothing on original clothing commodity graph.

The application provides a display panel and a display device, and solves the problem of crosstalk between adjacent OLED devices in a laminated OLED display panel in the prior art. The display panel comprises: a first light-emitting device layer; a second light-emitting device layer; and a charge generation layer between the first light-emitting device layer and the second light-emitting device layer. The display panel comprises a plurality of light-emitting areas and a spacing area between adjacent light-emitting areas; and the charge generation layer comprises a patterned structure layer, the patterned structure layer comprises at least one first electron generation unit and at least one first hole generation unit, and the at least one first electron generation unit and the at least one first hole generation unit are alternately arranged at least in the spacing area.

The application discloses a kind of multi-scale time sequence modeling video abstract generation method of fusion semantic enhancement and boundary perception, belong to computer vision technical field.The method includes: the video frame sequence of input is encoded, and the corresponding time sequence feature representation is extracted;The time sequence feature extracted is input video abstract generation model, and the importance score is output by model;The importance score is time sequence smoothing and peak detection, and the candidate set of key frame is determined;Based on the candidate set of key frame, in combination with non-maximum suppression and segment merging strategy, the final video abstract segment is generated.The experimental results obtained on benchmark dataset SumMe and TVSum prove the advancement of the method.The video abstract generation method proposed in the application can effectively solve the problems of event boundary ambiguity and insufficient multi-scale time sequence understanding in existing video abstract generation methods.

This application discloses a palladium-hydrogen rare earth nanoalloy material, its preparation method, and its application. The palladium-hydrogen rare earth nanoalloy material includes a carbon support and PdREH nanomaterials loaded on the surface of the carbon support. x Nanoparticles, PdREH x The nanoparticles comprise an alloy of Pd and rare earth elements (RE), and hydrogen embedded in the alloy lattice; wherein the rare earth element RE is one of Y, Ce, Sm, Gd, and Tb. This application constructs a nanoalloy material with a stable lattice structure and good hydrogen intercalation capability by forming an alloy of palladium and rare earth elements and further intercalating hydrogen. Simultaneously, by combining a carbon support with a two-step Joule thermal-hydrothermal preparation process, the composition and structure of the material can be controllably adjusted. This material exhibits excellent electrochemical response sensitivity, high ethylamine production capacity, and high Faradaic efficiency in the acidic electrocatalytic hydrogenation of acetonitrile to ethylamine reaction, making it suitable for low-resistance proton exchange membrane devices.

The application belongs to the field of artificial intelligence, computer vision and computer graphics, and discloses a method and system for VR real-time rendering of low-code-rate cloud rendering streaming, which comprises the following steps: obtaining a rendering frame image; constructing a lightweight super-resolution model; inputting the rendering frame image into the lightweight super-resolution model to obtain a reconstructed 4K resolution image; and performing real-time rendering and display on a VR device based on the reconstructed 4K resolution image. The application not only avoids the high bandwidth pressure caused by direct transmission of 4K video, but also overcomes the hardware limitation that low-config devices cannot perform complex rendering tasks. Through the combination of scene customization model and video coding, high-quality and low-delay visual experience is realized under limited resources, and the application is suitable for various application scenarios with high requirements for image quality and real-time performance, such as large-scale cloud simulation, VR tourism, intelligent power grid monitoring and the like.

The invention discloses a C3N5-C3N4 photocatalytic hydrogen evolution material as well as a preparation method and application thereof. The material consists of a core layer, a shell layer and an interface transition region, wherein the core layer is C3N5 rich in S-O bonds; the core layer is coated with the shell layer, the shell layer is C3N4 with gradient increasing nitrogen vacancy concentration from inside to outside, and CoP nanoparticles are loaded on the surface of the shell layer; and the C / N atomic ratio in the interface transition region is in continuous gradient change to form a built-in electric field gradient. The preparation method comprises the following steps: preparing a C3N5 core layer through pyrolysis of a sulfur-containing precursor; coating a C3N4 precursor on the surface through solvothermal reaction; carrying out heat treatment in a gradient ammonia atmosphere, carrying out synchronous crystallization and introducing gradient nitrogen vacancies; and finally, loading CoP through photodeposition. The material has spatially separated light absorption, charge transfer and reaction centers, shows excellent hydrogen evolution activity and stability under visible light, and is suitable for the field of photocatalytic hydrogen production.

This invention relates to the field of natural language processing technology, providing an end-to-end multi-turn dialogue rewriting method, system, and storage medium that integrates dialogue detection. In the proposed end-to-end multi-turn dialogue rewriting method, a joint learning approach is used to train the detector and generator, solving the problems of cascading errors and incoherent sentences after matching and filling in the Pipeline method, thus improving the detector's detection performance. By having the detector ignore sentences that do not need rewriting and integrating dialogue detection information into the generator, the slow speed and repetitive encoding problems of the Seq2Seq method are solved. The dialogue rewriting process of this invention does not require the generator to repeatedly encode the text, improving the model's prediction speed. Simultaneously, integrating dialogue detection information into the generator improves the generation quality of the rewritten sentences.