8 results about "Partial index" patented technology

The information processing apparatus includes at least one memory storing instructions and at least one processor that executes the instructions. The at least one processor executes instructions to acquire index information including a plurality of indexes related to a target medical institution, acquire an analysis result output by an analysis model that that has referred to at least a part of the index information, acquire explanatory information regarding at least a part of the analysis result, the explanatory information being information output from a generation model that has been subjected to machine learning and has referred to at least a part of the analysis result, and generate output data including at least a part of the analysis result and at least a part of the explanatory information.

This application discloses a method and system for processing local port numbers, a charging control terminal, and a computer-readable storage medium. The method for processing local port numbers includes: constructing a port index table, wherein the port index table includes multiple index values, and each index value corresponds to multiple local port numbers; allocating local port numbers to at least some of the index values ​​according to the index sequence number of the index values, and storing the allocated index values ​​and corresponding local port numbers in non-volatile memory; binding TCP sockets to each local port number stored in the non-volatile memory; and recording the bound local port numbers in a communication structure. This application improves port resource utilization efficiency while ensuring deterministic port allocation, and can restore the port allocation state after the device is powered off and restarted.

The invention relates to an index persistence and recovery method and device for elastic block storage and electronic equipment. The method comprises the steps that under the condition that a target read-write request is received, part of index entries, including a target index, in index entries persisted to a data disk are recovered into a memory, the target read-write request is used for requesting to read and write the target index in the memory, and the index entries persisted to the data disk are stored in the memory when the persistence condition is met. Persisting index entries to the data disk from the memory; obtaining a target index from a part of index entries; and returning the target index. The technical problem of low index searching efficiency is solved.

The invention relates to a permission rule storage and retrieval method which comprises an ordered index, a hash index and a directed acyclic graph index. The invention discloses an index-based authority rule management process, an addition process, a deletion process and a retrieval process. In the index-based authority rule management process, an index of a user range or an object range comprises two parts: an index information extractor for extracting index information from the user range or the object range; the index data manager is used for providing permission rule adding capability, permission rule deleting capability and permission rule query capability based on the index information extracted by the index information extractor; the two parts have a synergistic effect and jointly manage the authority rules, so that the problems of high time complexity and low efficiency of executing authority reasoning by a current authority rule storage and retrieval method are solved, and the method has lower retrieval delay and lower retrieval resource occupation.

The invention provides a data processing method and a data processing device. The data processing method comprises the steps of determining a target index corresponding to a current to-be-read data block according to a preset output sequence and at least part of indexes stored in an address memory, and reading and outputting the to-be-read data block from a target storage position, corresponding to the target index, of a data memory, and under the condition that the current data block to be written exists, storing the data block to be written at a target storage position of the data memory.

Embodiments of the present application disclose a database data row number counting method and device, equipment and a storage medium, comprising: when counting a row number of target data in a target database of an electronic device, acquiring an index range corresponding to the target data; wherein the target data is data meeting a counting condition in all to-be-counted data contained in the target database; generating a plurality of counting tasks according to the index range; wherein each counting task is used for counting the row number of the target data in a partial index range; distributing the plurality of counting tasks to at least one thread created in advance for processing to obtain a plurality of counting results; and adding the plurality of counting results to obtain a target counting result of the row number of the target data. Through implementation of the method, the counting speed of the row number of data meeting the counting condition can be improved for a large database with a data volume of ten million or more when the index cannot be compressed.

The application discloses a kind of multi-mode generative adversarial network modeling method and device for multidimensional sequence data.It includes normalizing original observation data, obtaining formal uniform standard observation data, organizing examples for multi-mode generative adversarial network model training based on standard observation data, designing the structure and training mode of basic module, establishing multi-mode generative adversarial network model and other steps.This model is suitable for learning sequence data with missing partial index data, reducing the waste of incomplete data samples.At the same time, the model is trained in the feature space, reducing the complexity of data dimension and model training, avoiding the occurrence of overfitting phenomenon.In the training of the model, the multiple possible feature data at the same position are constantly updated, increasing the potential examples available for training.Compared with the traditional generative adversarial network, a large number of positive samples are needed for model training.

The application discloses a story generation method and system based on external knowledge enhancement. The method is as follows: a knowledge base is constructed, including a knowledge set in a target field and a general knowledge set, each piece of knowledge in the knowledge base is a real story text; each piece of knowledge is divided into two parts: an index part and an extension part; a knowledge retrieval module retrieves matched indexes from the knowledge base according to an input prompt x, takes the extension part corresponding to each matched index as a candidate knowledge, and selects several keywords as external knowledge from each candidate knowledge; a knowledge selection module predicts a score for each selected keyword according to the input x and the candidate knowledge, and then selects knowledge for story generation according to the predicted scores of the keywords; a story generation module generates a story text y according to the selected knowledge for story generation; and then x and y are spliced into a coherent story.