7 results about "Denormalization" patented technology

The invention discloses a multi-scale time sequence prediction method based on an adaptive sparse expert selection strategy and a closed continuous time neural network. The method comprises the following steps: carrying out normalization and low-dimensional feature mapping based on RevIN; performing trend-seasonal structure enhancement processing on the feature sequence after linear mapping; constructing a multi-scale expert model based on the feature sequence after trend-season enhancement; self-adaptive sparse expert selection and load balancing loss calculation are carried out; carrying out weighted aggregation and residual fusion on multi-scale expert output; global modeling of a closed continuous time neural network based on channel weighting is carried out; and finally performing prediction generation and reverse normalization. The multi-scale time series prediction method has the multi-time-scale adaptive modeling capability, the sparse expert efficient selection mechanism and the global continuous time modeling capability, and can be applied to various multivariable time series prediction scenes such as power load prediction, weather prediction, industrial production monitoring, traffic flow prediction and financial price prediction.

The invention relates to the technical field of inventory management, and discloses an electricity coal inventory prediction method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining multi-source time series data related to the electricity coal inventory; aiming at a fluctuation rule of the multi-source time series data, carrying out adaptive learning to obtain a normalization parameter of a historical window and an anti-normalization parameter of a prediction window; the normalization parameters are utilized to carry out normalization transformation on the multi-source time sequence data, the normalized data are input into a preset multi-scale time sequence prediction model to obtain a preliminary prediction result, the multi-scale time sequence prediction model is used for extracting and fusing time sequence features of different time scales, prediction is carried out based on the fused multi-scale time sequence features, and the prediction result is obtained. Obtaining a preliminary prediction result; and performing anti-normalization on the preliminary prediction result by using the anti-normalization parameter to obtain a final electricity-coal inventory prediction value. According to the method, the data inherent rule is fully mined, the capturing capability of the model for complex time sequence changes is enhanced, and the accuracy of electricity-coal inventory prediction is improved.

The application discloses a container throughput prediction method based on stacked ensemble learning, and relates to the technical field of intelligent ports. In the system operation, multi-source data is collected from a port operation system, an economic statistics platform and a shipping database, a comprehensive feature system containing throughput, freight rate, transportation time, policy variable, seasonal characteristics and macroeconomic indicators is constructed, and serialization and standardization processing are performed, and an improved model is constructed. The improved CNN-LSTM model introduces a deep separable convolution, a bidirectional LSTM and an improved attention mechanism to enhance the representation ability of key time steps, adopts Bayesian optimization to automatically search for hyperparameters, combines an early stopping strategy to control overfitting, simultaneously realizes multi-model integration based on inverse error weighting and meta-learning linear regression model, adaptively adjusts the rolling prediction window size according to the data coefficient of variation, generates future multi-period throughput prediction results through multi-step rolling prediction, and performs denormalization output.

This invention discloses a maneuvering target tracking method based on a Mamba tracking network, comprising: 1. Preprocessing trajectory data, calculating the mean and standard deviation, and normalizing the input data to eliminate sequence non-stationarity; 2. Constructing a maneuvering factor generator to map statistical features into adjustment factors reflecting the target's maneuvering characteristics; 3. Constructing a target tracking network based on a Mamba tracking network, extracting long-range spatiotemporal dependence features of the trajectory using a selective scanning mechanism, and dynamically adjusting the network feature response through adjustment factors; 4. Performing denormalization on the network output to recover the target state prediction value at the physical scale. This invention, through the organic combination of the Mamba tracking network and adjustment factors, effectively compensates for the loss of maneuvering information during the normalization process, significantly improving the accuracy and generalization ability of tracking highly maneuvering targets while maintaining linear computational complexity, and has significant application value.

The invention discloses a data processing method, a computer readable storage medium and a computer program product, and relates to the field of integrated circuit design. The method comprises the steps that first data is determined based on input data, the input data is an index of an exponential function, and the first data is a decimal part of a fixed point number obtained by converting the input data; under the condition that the input data does not meet the normalization requirement, B-spline curve fitting is conducted on the first data, a first processing result is obtained, the normalization requirement is the requirement which needs to be met when the exponential function calculation result is a normalization number, and B-spline curve fitting is achieved through partial sum array calculation; performing normalization processing on the first processing result to obtain a second processing result; generating a non-normalized exponential function calculation result based on a preset index and the second processing result; wherein the index part of the non-normalized exponential function calculation result is a preset index, and the mantissa part of the non-normalized exponential function calculation result is a second processing result.

A system and computer implemented method for execution of aggregation expressions on a distributed non-relational database system is provided. According to one aspect, an aggregation operation may be provided that permits more complex operations using separate collections. For instance, it may be desirable to create a report from one collection using information grouped according to information stored in another collection. Such a capability may be provided within other conventional database systems, however, in a non-relational database system such as NoSQL, the system is not capable of performing server-side joins, such a capability may not be performed without denormalizing the attributes into each object that references it, or by performing application-level joins which is not efficient and leads to unnecessarily complex code within the application that interfaces with the NoSQL database system.

Systems and methods address the inefficiencies in document creation and management within large, globally dispersed organizations. The system leverages multiple independent cloud-based Generative AI (GenAI) models, each specialized and context-sensitive, to dynamically generate and contextualize documents tailored to specific regional and contextual requirements. A global repository meticulously catalogs and annotates documents with detailed metadata, enabling precise retrieval based on user-specific search criteria. The system employs a denormalization slant to generalize search queries, facilitating global context searches and affinity mapping to ensure relevance. A Primary Context Controller Hyper Model orchestrates the confederation of independent GenAI models, refining and enhancing documents to meet precise user needs while adhering to local nuances. Automated, rule-based approvers ensure instant validation and quality control, while a feedback loop mechanism continuously improves the models' accuracy and relevance. This decentralized, scalable approach reduces computational burden and costs, providing a robust, efficient solution for high-quality, context-aware document generation.