Abstract: Provided is a network modal management system and a management method. The system includes a polymorphic intelligent network integrated development environment and a polymorphic intelligent network distributed compilation and deployment environment. The polymorphic intelligent network integrated development environment provides users with an environment for writing code and configuration files, and facilitates users to debug the code. The polymorphic intelligent network distributed compilation and deployment environment provides users with a management interface and services, and integrates various functions related to network modal. The network modal management system significantly improves the management efficiency of polymorphic intelligent network platforms.

Abstract: The present application discloses a target detection method and device for a high-resolution remote sensing image, which comprises: acquiring an original high-resolution remote sensing image from a sensor; acquiring target information of an area and an expanded area of the area expanding around by a predetermined distance; adaptively partitioning the original high-resolution remote sensing image to obtain different cluster areas to be detected, and obtaining an area required to be detected and an area not required to be detected; selecting a model of the area required to be detected and generating a target detection scheme; executing the target detection scheme to obtain a detection result; determining whether a computing platform has extra computing resources to detect the area not required to be detected, if so, performing dynamic partition detection for the area to obtain a detection result, and merging the detection results into a target detection result.

Abstract: A method and a device for identification management and optimized forwarding in a large-scale polymorphic network, the method comprising the follow steps: S1, constructing a polymorphic backbone network; S2, modality identification management; S3, determining a modality to be forwarded; S4, configuring a flow table for a switching node; S5, receiving a packet by a balanced distributor, and preliminarily parsing the type of the packet; S6, parsing key field information in the packet, determining the switching nodes to be allocated according to the key field information, and transmitting the key field information to the corresponding switching node; S7, the switching node matching the stored flow table according to the key field information to determine a correct forwarding action.

Abstract: The present application disclose a method and a device for detecting and correcting abnormal scoring of peer reviews, which includes: converting collected scoring data into a two-dimensional matrix and preprocessing the data; determining the anomaly of the processed structured data with a one-way anomaly detection method, a consistency check method and a two-way anomaly detection method, and classifying the detected abnormal data into an abnormal data set; repairing the abnormal data for the abnormal data set with an information entropy correction method; generating an ability evaluation report according to the abnormal data set, performing weighed averaging on the corrected scoring data according to the scoring weights of reviewers in the ability evaluation report to obtain a final scoring result, and generating an abnormal scoring correction report. The present application can effectively detect the abnormal phenomenon of peer reviews in the performance appraisal of enterprise personnel.

Abstract: The present application discloses a system and a device for functional connectivity matrix processing based on feature selection using a filtering method, which comprises the following steps: acquiring a preprocessed resting state brain functional magnetic resonance image of a subject; extracting time series; calculating a Pearson correlation coefficient to obtain a Pearson correlation coefficient matrix; vectorizing the Pearson correlation coefficient matrix; calculating quantitative correlation indices using a filtering method, and selecting a quantitative correlation index based on a preset threshold; performing weighting processing a selected functional connectivity feature by using the corresponding quantitative correlation index with high correlation with a disease diagnosis result to obtain a functional connectivity matrix; and obtaining a prediction result from the functional connectivity matrix.

Abstract: The present disclosure relates to software-defined methods and apparatuses for designing a wafer-level switching system, including: determining wafer-level switching system layout constraints; constructing a target wafer-level switching system and determining parameters, and designing a logical topology of a switching network; designing a layout of the switching chiplets on the wafer substrate; respectively designing interface structures of external chiplets and internal chiplets; configuring a switching mode and an enable state of each port of the switching chiplets; ending the process when the target logical topology can be achieved by the wafer-level switching system; otherwise, reconstructing a logical topology of a switching network and mapping it to the substrate.

Abstract: A software and hardware collaborative compilation processing method and system. The system comprises an environment configurator, a command parser, a code filler, a scheduler and a heterogeneous target system, wherein the code filler is configured for obtaining the source program path of a user, reading source codes and identifying the heterogeneous target system according to a macro definition, complementing the codes related to the heterogeneous target system, carrying out primary filling and secondary filling on the source codes; the scheduler is configured for realizing compilation scheduling and execution scheduling functions respectively; the heterogeneous target system is configured for compiling and processing user modal data, and comprises at least two heterogeneous target subsystems; each target subsystem comprises a target-related middle-end compiler, a back-end compiler and a target-related running environment.

Abstract: The present disclosure discloses a reinforcement learning agent training method, modal bandwidth resource scheduling method and apparatus. The reinforcement learning agent training method utilizes a reinforcement learning agent to continuously interact with a network environment in a polymorphic smart network to obtain the latest global network characteristics and output updated actions. By adjusting the bandwidth occupied by modals, a reward value is set to determine an optimization target for the agent, the scheduling of modals is realized, and the rational use of polymorphic smart network resources is guaranteed. The trained reinforcement learning agent is applied to the modal bandwidth resource scheduling method, and can adapt to networks with different characteristics, and thus can be used for intelligent management and control of polymorphic smart networks and has good adaptability and scheduling performance.

Abstract: Disclosed is a method and an apparatus NER-orientated Chinese clinical text data augmentation, and unannotated data and annotated data of label linearization processing through data preprocessing. A concealed part is predicted based on retained information by using the unannotated data and concealing part of information in text, and meanwhile an entity word-level discrimination task is introduced for pre-training of a span-based language model; and a plurality of decoding mechanisms are introduced in a fine-tune stage, a relationship between a text vector and text data is obtained based on the pre-trained span-based language model, linearized data with entity labels is converted into the text vector, and text generation is performed through forward decoding and reverse decoding in a prediction stage of a text generation model to obtain enhanced data with annotation information.

Abstract: Disclosed a method and a system for overlapping sliding window segmentation of an image based on an FPGA. According to the method, on-chip BRAMs storage resource cost of FPGA is determined; each on-chip BRAM in FPGA is used to cache the pixel data of each segmented sub-image in parallel; when the pixel data received by the BRAMs reaches a preset threshold or the last pixel of the segmented sub-image is written into the on-chip BRAMs, the data is written from the on-chip BRAMs to an off-chip DDR memory in a burst continuous writing mode; the repeated data generated by segmentation of horizontally overlapping sliding windows are written into the on-chip BRAMs corresponding to the current segmented sub-image and adjacent segmented sub-images thereof respectively in a synchronous and parallel manner.

Abstract: The present disclosure discloses Spark collaborative computing, job method and apparatus for multiple K8s clusters, and addresses the problem that most of the current multiple K8s clusters adopt the model of federated clusters, and Spark's own method of scheduling and optimization cannot be implemented across domains, by implementing a cross-domain job method, setting the multiple K8s clusters as a master cluster and a slave cluster, with the master cluster being responsible for creating Spark's The master cluster is responsible for creating Spark's Driver container and Pods, and the slave cluster is responsible for creating Spark's Executor container and Pods. After the containers are created, a direct tunnel is established between the master cluster and the slave cluster by aggregating address information and access credentials through the Collaboration Center, and the containers in the slave cluster register with Driver and continuously send heartbeat messages through the tunnel.

Abstract: The present disclosure belongs to the field of intelligent computing technologies, and relates to a multi-policy intelligent scheduling methods and apparatuses oriented to heterogeneous computing power. The method includes: step 1, setting an execution policy of a task based on heterogeneity of computing clusters, differences of computing tasks and a user requirement, and constructing a Markov decision process model by adopting a reinforcement learning method combined with the execution policy; step 2, adopting a proximal policy optimization to solve an optimal task scheduling policy of the task input by the user based on the constructed Markov decision process model; step 3, scheduling the task to a corresponding computing cluster for execution based on the optimal task scheduling policy.

Abstract: The present disclosure discloses a method for execution of a computational graph in a neural network model and an apparatus thereof, including: creating task execution bodies on a native machine according to a physical computational graph compiled and generated by a deep learning framework, and designing a solution for allocating a plurality of idle memory blocks to each task execution body, so that the entire computational graph participates in deep learning training tasks of different batches of data in a pipelining and parallelizing manner.

Abstract: Disclosed are a data flow method and apparatus for neural network computation. The data flow method for neural network computation includes initializing the lifecycle of a variable in a computational graph; and defining a propagation rule for a variable in use to flow through a node. A definition of the variable is produced at a precursor node of the node, such that an input set of valid variables flowing through the node contains the variable. The method may be used on neural network computation in a deep learning training system.

Abstract: Disclosed is a method for adapting a deep learning framework to a hardware device based on a unified backend engine, which comprises the following steps: S1, adding the unified backend engine to the deep learning framework; S2, adding the unified backend engine to the hardware device; S3, converting a computational graph, wherein the computational graph compiled and generated by the deep learning framework is converted into an intermediate representation of the unified backend engine; S4, compiling the intermediate representation, wherein the unified backend engine compiles the intermediate representation on the hardware device to generate an executable object; S5, running the executable object, wherein the deep learning framework runs the executable object on the hardware device; S6: managing memory of the unified backend engine.

Abstract: The present application discloses a traffic light control method for an urban road network based on expected return estimation, which uses C-V2X wireless communication technology to obtain real-time information of all vehicles and traffic state in the road network from vehicle-mounted terminals, and adaptively and dynamically controls the phase transformation of the traffic light. According to the present application, the expected returns of keeping the current phase and executing phase switch are calculated by estimating the timely driving distance and the future driving distance of the passable vehicles in the next green light duration in combination with the proposed road priority traffic index. By comparing the expected returns of keeping the current phase or switching to other phases, the best phase is selected, so as to make as many passable vehicles travel farther as possible in the next green light duration. Therefore, the efficiency of traffic will be improved.

Abstract: The present disclosure discloses a distributed model compilation system. A master node of the system determines the logic calculation graph of the model based on model information, divides the logic calculation graph into multiple logic calculation sub-graphs, generates a distributing message for each logic calculation sub-graph, and then transmits the distributing message to a slave node. Each of the slave nodes allocates a local computing resource to compile the logic calculation sub-graph based on the received distributing message, and transmits compilation completion information to the master node. The master node determines the completion of model compilation based on the compilation completion information returned by each slave node, and executes the target work based on the compiled model.

Abstract: The present invention discloses a two-dimensional photonic neural network convolutional acceleration chip based on series connection structure, which is integrated with a modulator, M microring delay weighting units, M?1 secondary delay waveguide, a wavelength-division multiplexer and a photodetector.

Abstract: The disclosure discloses a mimetic database-based network operating system design method, including: designing a mimetic data structure; designing a mimetic data object; designing a synchronization mechanism and a decision mechanism, designing a mimetic database safe storage command processing system, and designing a classification storage mechanism for interacting data between service modules and a master database in a network operating system. By means of vertical hierarchy and horizontal classification, the problem of compatibility of the database subjected to mimetic transformation and a network operating system is solved. By means of a memory random distribution storage mechanism and a memory hardware heterogeneous storage mechanism, the cost caused by mimetic transformation can be reduced, and the cost is controllable while the safety is improved.

Abstract: A polymorphic network control system and a polymorphic network control system method supporting terminal mobile access. A network architecture is redesigned based on the idea of separation of an identifier from a locator. In this network mode, the three-layer protocol is no longer a traditional IP message, but a message carrying a locator and an identifier. The polymorphic SDN network controller is responsible for forwarding the data packet to a destination corresponding to a destination locator. When the network location of a mobile device is changed, the transmission connection established based on the identifier will suspend communication due to the disconnection of the physical link. When the mobile device is re-accessed and a new forwarding path is established, the transmission connection can recover communication.