6 results about "Predictive regression" patented technology

The application discloses an agent action prediction method based on multi-scale space perception, comprising map space structure modeling, historical trajectory feature extraction, feature fusion and multi-modal action prediction, wherein: the map space structure modeling uses a multi-scale graph convolutional neural network to extract map features from two-dimensional vector map data in an application scenario map, to obtain high-dimensional map feature information; the historical trajectory feature extraction uses a convolutional neural network and a feature pyramid network to extract high-dimensional trajectory data features of all agents; the feature fusion models and fuses the correlation of high-dimensional map feature information and high-dimensional trajectory data feature information through a self-attention mechanism, to obtain agent trajectory fusion features with direction information; and the multi-modal trajectory prediction uses the agent trajectory fusion features for prediction regression and confidence scoring, to output multi-modal complete trajectory coordinates and corresponding confidence scores of action prediction, so as to provide reasonable auxiliary decision-making for agent action.

This disclosure relates to the field of bridge engineering, specifically to a design method, system, and medium for HPC-RC combined eccentrically compressed columns. The method includes: defining a design space; selecting optimal design data combinations based on engineering specification constraints, bearing capacity constraints, and cost-effectiveness values ​​to construct a design data sample set; constructing a two-branch heterogeneous neural network model, the model including an input layer, a shared feature extraction layer, a diameter prediction classification branch, and a reinforcement area prediction regression branch; constructing a loss function composed of focal loss and mean square error loss; training the two-branch heterogeneous neural network model using the design data sample set; and using the trained two-branch heterogeneous neural network model to predict the diameter and total reinforcement area of ​​the eccentrically compressed column. This disclosure improves the stability and accuracy of the design, reduces the computational burden, increases computational speed, and reduces computational complexity.

The invention relates to the technical field of disease risk intelligent prediction, in particular to a chronic kidney disease progress risk prediction system based on machine learning, which comprises an accumulated impact acquisition module used for inputting disease information and physiological index data of a current patient into a disease risk transmission network to acquire an accumulated impact condition; the predicted eGFR calculation module is used for inputting the accumulated impact condition into an eGFR prediction regression network to obtain a predicted eGFR curve; the similarity analysis module is used for calculating a similarity measurement value between the eGFR curves of the current patient and each target historical patient; the risk value calculation module is used for obtaining a renal function attenuation risk value based on the predicted eGFR curve, the similarity measurement value and the eGFR difference value between the current patient and each target historical patient; and the risk judgment module is used for judging whether the current patient has a renal function collapse risk based on the renal function attenuation risk value. According to the invention, early and accurate prediction can be carried out on the CKD progress risk.

The present disclosure relates to the field of bridge engineering, and particularly relates to a design method, system and medium for HPC-RC combined eccentric compression column. The method comprises the following steps: defining a design space; screening out an optimal design data combination according to engineering specification constraints, bearing capacity constraints and performance-price ratio value, and constructing a design data sample set; constructing a double-branch heterogeneous neural network model, wherein the model comprises an input layer, a shared feature extraction layer, a diameter prediction classification branch and a reinforcement area prediction regression branch; constructing a loss function composed of focal loss and mean square error loss, and training the double-branch heterogeneous neural network model by using the design data sample set; and predicting the diameter of the eccentric compression column and the total reinforcement area of the eccentric compression column by using the trained double-branch heterogeneous neural network model. The present disclosure improves the stability and accuracy of the design, has a small calculation burden, improves the calculation speed and reduces the calculation amount.

This invention relates to the field of building structural performance evaluation, specifically a machine learning-based method for evaluating the seismic resistance of unreinforced masonry buildings. The method includes considering the spatial distribution and cumulative overall evaluation index of damage; pre-setting different levels of damage states; using an adaptive multi-scale progressive nonlinear dynamic response analysis method to obtain the median critical peak acceleration when the index reaches the limit value of each level under different working conditions; training a seismic resistance prediction regression model using the median values ​​of different levels; training a seismic resistance classification model using the median values ​​of each level; fitting adjustment parameters for service life; using the prediction model to obtain the median value prediction results corresponding to each level during evaluation; and inputting the prediction results optimized with the adjustment parameters into the classification model to obtain the building grade. This invention solves the technical problems of high computational cost of traditional evaluation methods, insufficient sample size and poor interpretability of machine learning methods, and difficulty in practical application of existing evaluation methods.

This invention belongs to the field of computer vision and pattern recognition technology, specifically relating to a method and system for predicting energy consumption during fitness activities based on vision and deep learning. It aims to address the problem that existing vision-based general energy consumption prediction models suffer from severe feature interference and limited prediction accuracy due to neglecting biomechanical differences between different movement patterns. The invention includes: acquiring video stream data and extracting key human body points to construct a skeleton sequence tensor; identifying movement pattern categories using a spatiotemporal graph convolutional network; extracting associated kinematic feature time series based on movement categories; inputting the features into an energy consumption prediction regression model with an independent parameter space, encoding temporal dependencies based on a self-attention mechanism, and outputting predicted values ​​of body activity intensity. This invention, through a hierarchical architecture of identification followed by prediction, effectively improves the accuracy and robustness of energy consumption prediction across movement patterns.