Modeling system and method for visual machine learning training model
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A machine learning and training model technology, applied in machine learning, computing models, instruments, etc., can solve the problems of data analysts and engineers spending a lot of time, model training time-consuming, time-consuming and other problems, and achieve the effect of speeding up training efficiency
Pending Publication Date: 2017-09-15
北京天机数测数据科技有限公司
View PDF1 Cites 65 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0003] Among them, each stage needs to be coded independently, especially the creation and analysis process is very cumbersome and time-consuming, requiring data analysts and engineers to invest a lot of time
[0004] In addition, due to the non-uniform format of exchanged data at each stage, model training is very time-consuming and systematic result verification cannot be achieved
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
example 1
[0102] Example 1 (text mining, one million pieces of text data)
[0103] Its analysis process includes:
[0104] 1. Text preparation;
[0105] 2. Stop word filtering;
[0106]3. Word frequency statistics;
[0107] 4. Feature extraction;
[0108] 5. Model training with logistic regression algorithm;
[0109] 6. Evaluate logistic regression;
[0110] 7. Export the model after strong training.
[0111] When faced with this example in the prior art, it needs to be completed step by step independently, each stage requires programming, and the overall training time needs 3 hours.
[0112] With the system and method of the present invention, based on visual process design, model verification, and distributed computing, the overall training time only needs half an hour, and compared with the prior art, the efficiency is significantly improved.
example 2
[0113] Example 2 (weather analysis)
[0114] Construct a model for evaluating wind energy resources, by analyzing historical data of wind resources (daily and annual changes of wind speed and wind power density at each height in the last 22 years and their average conditions, frequency distribution of wind speed at different heights, wind direction frequency and direction of wind energy density distribution, wind speed and wind energy frequency distribution, annual effective wind speed hours, turbulence, wind shear index, air density...), predict the availability of wind resources in the next 5 years.
[0115] With the prior art, it takes one person two days to complete the creation and export of the model, but with the present invention, it only takes one hour to complete the creation and export of the model.
[0116] In summary, the present invention can realize visualized process design, visualized model verification, and visualized intermediate result viewing, allowing dat...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The invention relates to a modeling system and method for a visual machine learning training model. The modeling system for a visual machine learning training model includes a process designer, a process analyzer, and a process scheduler, wherein the process designer is used for according to dragging operation of the graphic algorithm component selected by a user, establishing the data flow direction between the algorithms in the graphic algorithm component, and generating the process description language; the process analyzer is used for analyzing the process description language generated from the process designer, creating the corresponding learning component, and generating the corresponding Spark learning pipeline; and the process scheduler is used for submitting the Spark learning pipeline to a Spark cluster to perform model training. By selecting the corresponding graphic algorithm component, establishing the data flow direction between the algorithms through dragging, generating the process description language, analyzing the process description language, creating the corresponding learning components according to the node class name and the attribute, generating the corresponding Spark learning pipeline, and submitting the Spark learning pipeline to the Spark cluster for performing model training, the modeling system and method for a visual machine learning training model can realize high quality machine learning modeling.
Description
technical field [0001] The invention belongs to the technical field of big data machine learning, and in particular relates to a visualized machine learning trainer, which is mainly used to help users realize fast model training. Background technique [0002] The creation process of existing machine learning models is very cumbersome, and its creation process usually includes: feature analysis, model training, model verification, model tuning, model export and model loading. [0003] Among them, each stage needs to be coded independently, especially the creation and analysis process is very cumbersome and time-consuming, requiring data analysts and engineers to invest a lot of time. [0004] In addition, due to the non-uniform format of exchanged data at each stage, model training is very time-consuming and systematic result verification cannot be achieved. Contents of the invention [0005] In order to solve the above-mentioned problems in the prior art, the present inve...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more