5results about How to "Optimize execution efficiency" patented technology

This invention, belonging to the field of computational processing technology, proposes a method to accelerate sparse 3D convolution based on thread bundle alignment and memory access rearrangement. To address the issues of discontinuous memory access and low memory bandwidth utilization caused by the sparsity of convolution kernels in GPU architectures, this invention establishes a thread bundle aligned convolution kernel format method based on the direct convolution method. This method uses the width of the thread bundle as the basic unit for grouping and compression to obtain the thread bundle aligned convolution kernel format. Considering the single-instruction multithreading (SIMT) characteristics of graphics processors and the properties of convolution operations, this method is used to merge sparse convolution kernel data and perform vectorized memory access. A memory access conflict resolution rearrangement method is also established: Input data is stored in shared memory using a DHWC layout, and then a greedy algorithm is used to rearrange the weight storage order within each basic unit of the WAF, thereby eliminating shared memory conflicts for input data access by manipulating the address of each memory access.

The present application relates to a kind of rail transit line congestion train operation control system and method, comprising: step 1, according to the delay time of congestion, determine the each train in the range of affected area;Step 2, introduce travel speed-interval calculation model, obtain the slow-down multi-stop control time of each train for each train in the range of affected area;Step 3, slow-down multi-stop control time is automatically issued to each train to execute.The system includes intelligent scheduling strategy operation module and intelligent scheduling strategy decomposition execution module, for running travel speed-interval calculation model, dynamically calculate the slow-down multi-stop control time of each train, and output to each train for execution.The system relies on algorithm to automatically control the automatic adjustment of all trains in upstream and downstream direction, without scheduling manual control, and execution efficiency is excellent.

The invention belongs to the technical field of terminal testing, and particularly relates to a mobile intelligent terminal compatibility testing method which comprises the following steps: S1, aggregating terminal hardware configuration, software environment and network state data through federal learning, and constructing a real-time testing matrix; s2, synthesizing an edge scene test case based on the generative adversarial network; s3, synchronously collecting a touch track, a display frame rate and an audio synchronization deviation, and quantifying user interaction experience in combination with eye tracker data; s4, deploying a lightweight virtualization intermediate layer between the gap / Android / iOS, automatically converting an equipment interconnection instruction, and verifying cross-chain compatibility; and S5, tracking the privacy data flow by adopting dynamic taint analysis, and detecting illegal behaviors in real time. According to the method, through dynamic environment modeling, real-time test matrix construction and hierarchical aggregation architecture, terminal nodes encrypt and upload local data by using differential privacy and homomorphic encryption technologies, so that the security of the data is ensured.

The application discloses a multi-model reasoning method and device based on a graph structure, a medium and a program product, wherein the method comprises the following steps: querying a task target node in a graph database and starting from the task target node, and performing reasoning path expansion in the graph database according to a corresponding driving model, wherein the graph database comprises a plurality of directed acyclic graphs, and each directed acyclic graph is obtained by mapping metadata of a registered analysis model; generating a task topology execution graph according to model nodes involved in the expanded reasoning path and the corresponding driving model; and sequentially scheduling actual execution of each model node according to the task topology execution graph, and obtaining an explanation result corresponding to the task target node. The application improves the reasoning execution efficiency by explicitly attributing the analysis model and optimizing the reasoning path by using graph calculation, and improves the intelligence, performance and interpretability of multi-model collaborative analysis.

The application provides a kind of multi-model API docking management method and system based on deep learning, it is related to artificial intelligence and cloud computing technology field, including the following steps: collect and standardize all the AI prompt word templates and skill assets generated in workflow storage, to generate AI prompt word template library and skill asset library;Request data is optimized;The state of the multi-business model is evaluated in combination with the optimized request data to determine the target business model, and the optimized request data is distributed to the target business model to perform inference operations;Real-time acquisition of inference results and performance indicators generated during the execution of the target business model, and the inference results and performance indicators are returned to the cloud storage space;The AI prompt word template library and skill asset library are updated to realize the docking optimization of multi-business model.The application realizes the efficient, low-cost and adaptive optimization of multi-business model call, with good scalability and practical value.