42results about How to "Improve model quality" patented technology

A system and method for assembling a mesh model for engineering analysis by a user is provided. The system includes a user computer system, a remotely located computer system and a data storage means. The method includes the steps of selecting a plurality of component parts having a best practice mesh model format and selecting a predetermined property for each of the selected component parts, wherein the predetermined property is selected from a best practice library stored on the data storage means. The method also includes the steps of determining if the mesh model for each of the selected component parts meets a predetermined mesh quality condition using a best practice mesh quality checking software program. The method further includes the steps of selecting a best practice connecting means for connecting the selected component parts, wherein the connecting means is selected from a best practice library of mesh models of connecting means stored in the data storage device, and assembling the selected component parts together using the connecting means and the best practice properties into a new mesh model using a best practice mesh assembly software program, and the assembled mesh model is available for further analysis.

A computer method and apparatus of online automated model identification of multivariable processes is disclosed. The method and apparatus carries out automatically all the four basic steps of industrial process identification: 1) identification test signal design and generation, 2) identification plant test, 3) model identification and 4) model validation. During the automated plant test, process models will be automatically generated at a given time interval, for example, every hour, or on demand; the ongoing test can be automatically adjusted to meet the process constraints and to improve the data quality. Plant test can be in open loop operation, closed-loop operations or partly open loop and partly closed-loop. In a (partial) closed-loop plant test, any type of controller can be used which include proportional-integral-derivative (PID) controllers and any industrial model predictive controller (MPC). The obtained process models can be used as the model in advanced process controllers such as model predictive control (MPC) and linear robust control; they can also be used as inferential models or soft sensors in prediction product qualities. The apparatus can be used in new MPC controller commissioning as well as in MPC controller maintenance.

A system for reviewing role definitions includes a database that stores a plurality of roles. Each of the plurality of roles associates an identifier with at least one entitlement. The system also includes a second database that stores role review definitions. The review definitions define role review processes for the plurality of roles. The system further includes a processor that is in communication with the role database and the review definition database. The processor executes the role review processes defined by the review definitions at the appropriate times.

Systems and methods involving distributed machine learning using fog computing are described. The distributed machine learning architecture described involves at least a cloud server, one or more fog nodes and one or more edge devices. The cloud server has superior computational power compared to the fog nodes and edge devices and the edge devices may have inferior computational power compared to the fog nodes. The cloud server, fog nodes and edge devices may each have machine learning capability involving learning algorithms used to train models that may be used for inferencing. The distributed machine learning platform described herein may be used for making predictions and identifying certain types of data or trends in data. By distributing the machine learning computation to lower level devices, such as fog nodes and edge devices, bandwidth usage and latency common in traditional distributed systems may be reduced.

The invention discloses a federation learning modeling method, a device and a readable storage medium. The federation learning modeling method comprises the steps of carrying out federation learning modeling, receiving encryption model parameters sent by each second device and verification parameters corresponding to the encryption model parameters, based on each verification parameter, obtaininga verification result; and performing zero knowledge verification on each encryption model parameter to determine a false encryption model parameter in each encryption model parameter to obtain a zeroknowledge verification result, and then coordinating each second device to perform federation learning modeling based on the zero knowledge verification result and each encryption model parameter. The technical problems that federation learning modeling is low in efficiency and poor in accuracy are solved.

The invention discloses an automatic three-dimensional modeling method based on a contour line of an ore body. The automatic three-dimensional modeling method comprises obtaining a contour line of an ore body, cutting the ore body into Q planes, and obtaining Q contour line sets, where Q is a preset positive integer and the planes are prospecting line profiles or midplanes; obtaining volume data of the ore body according to the contour line sets; extracting a contour surface of the ore body according to the volume data; and generating an ore body three-dimensional model according to the contour surface. The automatic three-dimensional modeling method based on the contour line of the ore body is used for modeling of an ore body with an outer surface contour, and establishing an ore body model automatically through the contour line of the ore body. The method has the characteristics of high computer automation degree, high robustness, high modeling quality and the like.

A head device of three-dimensional modeling equipment is disclosed which has unidirectionally rotating polygon mirrors, can perform biaxial scanning at a high speed due to a combination of the mirrors, can easily control timing and a modeling ray irradiation position, and can enhance modeling precision. A scanning method for a modeling plane using the same is also disclosed.

The invention discloses a graph model intelligent commodity recommendation method fusing a knowledge graph and user interaction, and the method comprises the steps: 1, collecting the historical interaction record data of a user for a commodity, constructing a user commodity interaction matrix Y for training a recommendation model, and constructing a user commodity interaction bipartite graph; 2, collecting commodity attribute features and association features between attributes, and constructing a knowledge graph by using priori knowledge; 3, constructing a recommendation model fusing the knowledge graph and user interaction, and selecting a proper loss function to optimize model parameters and feature vectors; and 4, predicting the probability that the user interacts with the non-interacted commodities in the future by using the recommendation model, and selecting the commodity with the maximum interaction probability to recommend to the user, thereby completing a commodity recommendation task. According to the method, graph convolution operation on the knowledge graph and the interactive bipartite graph is combined, and semantic and structural information carried by the knowledge graph can be more sufficiently captured, so that a more accurate recommendation effect is realized.

The embodiment of the invention discloses a model training method, a face gender prediction method, equipment and a storage medium, and belongs to the field of face gender prediction. The method comprises the following steps: acquiring a sample face image; performing face correction on the sample face image to obtain a sample face image after face correction; performing data enhancement processingon the corrected sample face image to generate at least one target sample face image corresponding to the sample face image; and training a face gender prediction model according to the target sampleface image and the gender label of the sample face image. By adopting the method provided by the embodiment of the invention, the face correction is performed on the sample face image, so that the influence of non-uniform face orientations on the subsequent model training is avoided, and the model quality of the face gender prediction model obtained by training is further improved; and a sample face in a sample face image is transformed by adopting a data enhancement mode, so that the sample data volume can be increased, and the workload of sample face image acquisition is reduced.

A process implemented as software for building, developing, and enhancing a model for use in forecasting, having a first user input step, wherein a user to input data using a user interface on a user device and providing the user input data to an application program interface (API); the API performs an auto data validation step, a feature creation step comprising using domain knowledge to extract features from raw training data; a feature encoding step comprising using the created features and raw training data to train different candidate models; a model selection step wherein the user is prompted to select a best model from the number of trained candidate models based on user defined model rankings; a best model review step comprising producing detailed information on the best model through statistical diagnostics, sensitivity, back-test and performance analysis; and generating implementation code for the best model; processing a set of data to be analyzed using the best model, forecasting an outcome based on processing the set of data to be analyzed with the best model, and providing the forecast to a user by a user interface on a user device.

The invention discloses a dislocation equalization joint eliminating method and system for a large-breadth DLP type 3D printing machine. The method includes the steps that a three-dimensional model isevenly sectioned, and a section bitmap is generated; dislocation segmentation is conducted on the section bitmap, and a segmentation bitmap is generated; and a moving projector is used for conductingmoving dislocation splicing on the segmentation bitmap, and a three-dimensional model solid is generated. By means of the method, the forming size of the 3D printing machine can be enlarged with small additional cost, the joint problem is well solved, the overall quality of the model is improved, meanwhile, hardware complexity is lowered, and very good expandability is achieved.

The invention relates to the field of baijiu year prediction, in particular to a method for predicting the storage year of baijiu and aims to rapidly and accurately predict the storage year of baijiu. The method for predicting the storage year of Baijiu comprises the following steps of: acquiring volatile flavor component fingerprint spectrum of Baijiu with different storage time by adopting GC-MS; screening out modeling features from the fingerprint spectrum through extreme random forest regression and sklearn features; taking the modeling features as features of an XGboost regression model to establish a prediction model; and finally predicting the storage year of the baijiu through the prediction model. The method is suitable for rapidly and accurately predicting the storage year of the baijiu.

The invention discloses a text filtering method and device and a computer storage medium. The method comprises the following steps: obtaining text fluency based on a language model; obtaining an effective word rate based on an effective word dictionary constructed in a user-defined manner; when the text fluency meets a first preset threshold value and the effective word rate meets a second preset threshold value, executing a filtering operation on the text. According to the method, the problems of time and labor consumption, low efficiency, high cost and low quality of manual text screening and filtering are solved, and the semantic-level and character-level text screening quality in the corpus is improved, so the training model and service quality are improved, and the calculation overhead is reduced.

The invention provides a positioning reconstruction method and device. The positioning reconstruction method comprises the following steps: acquiring at least two frames of images including a target object; calculating the optical flow of the target object in the at least two frames of images; calculating attitude information of an image acquisition device for acquiring the image according to theoptical flow of the target object; calculating the position information of the target object and the confidence coefficient of the position information according to the optical flow of the target object and the attitude information of the image acquisition device; and establishing a model of the target object in a map based on the position information and the confidence of the position information. According to the positioning reconstruction method, uncertain factors in a position environment can be more effectively dealt with, and the advantages of data fusion are brought into full play, andall information is reserved through the probability density function, and the data is prevented from being discarded, so that the accuracy of positioning and modeling is improved, and the modeling quality is improved.

A system, method and memory medium for injecting positional information into a geological data set for improved modeling of surfaces in a geological formation. The method involves: (1) displaying a cross section of the formation through a selected subset of wells, where the displayed cross section includes well picks for the selected wells; (2) opening a gap in an unconformity of the displayed cross section, where the gap is opened at one or more wells that intersect the unconformity, where the gap represents non-deposition and/or material that is missing from the formation due to erosion; (3) receiving user input specifying positions of tops within the gap; and (4) closing the gap, where closing the gap includes calculating positions of restored tops based on the positions of the tops specified within the gap, where the positions of the restored tops are usable to model one or more surfaces in the formation.

The invention discloses a commodity recommendation method based on meta learning and a knowledge graph, and the method comprises the steps: 1) obtaining the historical interaction record data of a user commodity, and constructing a user-commodity interaction bipartite graph as the training data of a recommendation model; 2) obtaining commodity attributes and association features among the attributes, and constructing a background knowledge graph by using priori knowledge; 3) constructing a recommendation model based on meta learning and a knowledge graph, and selecting a proper loss function to optimize model parameters and feature vectors; and 4) using the recommendation model to carry out score prediction on the possibility that the non-interacted commodities generate interaction in the future, and finally obtaining a commodity recommendation result according to a score sequence. According to the method, the technology based on meta-learning and the knowledge graph is adopted, the background knowledge graph of commodity attributes is fused, and commodities which the user may be interested in can be accurately recommended for the user only by using a small amount of training data to train a recommendation system.