CN109583568A - Data extension method, device and electronic equipment - Google Patents
Data extension method, device and electronic equipment Download PDFInfo
- Publication number
- CN109583568A CN109583568A CN201811453086.8A CN201811453086A CN109583568A CN 109583568 A CN109583568 A CN 109583568A CN 201811453086 A CN201811453086 A CN 201811453086A CN 109583568 A CN109583568 A CN 109583568A
- Authority
- CN
- China
- Prior art keywords
- neural network
- rank
- extension
- network
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of data extension method, device and electronic equipments, which comprises original observed image data is divided into N number of rank, obtains classification image document corresponding to current time each rank；Extended using deep neural network based on the classification image document of each rank, obtain extension data corresponding to each rank of subsequent time, for deep neural network using the weight coefficient of preset tactful neural network output as input, weight coefficient makes the network state Maximum Value of deep neural network；Image synthesis is carried out to extension data corresponding to N number of rank, reconstruct the extension observed image data of subsequent time, reaching can be using the weight coefficient configurable deep neural network for the network state Maximum Value for making deep neural network, and utilize the extension data extended under residual error minimum with the deep neural network postponed, the extension observed image data of subsequent time is finally reconstructed, the technical effect for extending accuracy is improved.
Description
Technical field
The present invention relates to data elongation technology fields, more particularly, to a kind of data extension method, device and electronic equipment.
Background technique
The advantage of existing forecast system mainly uses the methods of machine learning and intensified learning, establishes analysis system
System strengthens the subsystems such as forecast system and forecast analysis system, relative to original single Numerical Prediction Models, to the full extent
Enhance the function of forecast system.Analysis system increases the accuracy of numerical model initial fields, strengthens forecast system and increases
The accuracy of Forecast Mode is added, forecast analysis system increases the accuracy of forecast data.
However, the uncertainty of the initial fields and Numerical Prediction Models itself due to Numerical Prediction Models, existing pre- syndicate
Itself error of uniting is larger, and forecast data accuracy is not high.
Summary of the invention
In view of this, the purpose of the present invention is to provide a kind of data extension method, device and electronic equipment, it is existing to alleviate
The technical problem for having error present in technology larger.
In a first aspect, the embodiment of the invention provides a kind of data extension methods, comprising:
Original observed image data is divided into N number of rank according to image attributes, is obtained corresponding to current time each rank
Classification image document, wherein N be more than or equal to 1；
Extended using preset deep neural network based on the classification image document of each rank, is obtained next
Extension data corresponding to moment each rank, the weight system that the deep neural network is exported with preset tactful neural network
Number makes the network state Maximum Value of the deep neural network as input, the weight coefficient；
Image synthesis is carried out to extension data corresponding to N number of rank, reconstructs the extension observed image money of subsequent time
Material.
With reference to first aspect, the embodiment of the invention provides the first possible embodiments of first aspect, wherein institute
Stating tactful neural network includes: value assessment network and control neural network, and the value assessment network is for assessing the depth
The value of the network state of neural network is spent, value that the control neural network export based on the value assessment network and in advance
If adjustable strategies adjust the weight coefficient.
With reference to first aspect, the embodiment of the invention provides second of possible embodiments of first aspect, wherein institute
State weight coefficient described in value adjustment of the tactful neural network for exporting based on preset value assessment network, comprising:
When the value increases, forward direction adjusts the weight coefficient of the deep neural network, makes the depth nerve net
The value of the network state of network increases；
If the value reduces, the weight coefficient of the neural network is reversely adjusted, the deep neural network is made
The value of network state reduces.
With reference to first aspect, the embodiment of the invention provides the third possible embodiments of first aspect, wherein institute
The default adjustable strategies for stating control neural network are that RNN clustered control network is sent, and the RNN clustered control network is used for base
In value assessment network output value to the control neural network output adjustment strategy.
With reference to first aspect, the embodiment of the invention provides the 4th kind of possible embodiments of first aspect, wherein institute
Stating deep neural network includes multiple CNN neural networks, the classification image document of each the CNN neural network and a rank
It is corresponding, extended for the classification image document based on N grades；The strategy neural network and the CNN neural network one
One is corresponding, and the strategy neural network and the RNN clustered control network correspond.
With reference to first aspect, the embodiment of the invention provides the 5th kind of possible embodiments of first aspect, wherein institute
It states and original observed image data is divided into N number of rank according to image attributes, obtain classification corresponding to current time each rank
Image document, comprising:
Obtain original observed image data；
The original observed image data is subjected to Wavelet image decomposition, obtains dividing corresponding to current time each rank
Grade image document.
With reference to first aspect, the embodiment of the invention provides the 6th kind of possible embodiments of first aspect, wherein institute
It states and image synthesis is carried out to extension data corresponding to N number of rank, reconstruct the extension observed image data of subsequent time, wrap
It includes:
Image synthesis is carried out to extension data corresponding to N number of rank using wavelet inverse transformation, reconstructs subsequent time
Extend observed image data.
Second aspect, the embodiment of the present invention also provide a kind of data extension apparatus, comprising:
It is every to obtain current time for original observed image data to be divided into N number of rank according to image attributes for diversity module
Classification image document corresponding to a rank, wherein N is more than or equal to 1；
Extension of module, for being carried out using preset deep neural network based on the classification image document of each rank
Extend, obtains extension data corresponding to each rank of subsequent time, the deep neural network is with preset tactful nerve net
For the weight coefficient of network output as input, the weight coefficient makes the network state Maximum Value of the deep neural network；
Synthesis module reconstructs prolonging for subsequent time for carrying out image synthesis to extension data corresponding to N number of rank
Stretch observed image data.
The third aspect, the embodiment of the present invention also provide a kind of electronic equipment, including memory, processor, the memory
In be stored with the computer program that can be run on the processor, the processor is realized when executing the computer program
The step of stating method described in first aspect.
Fourth aspect, the embodiment of the present invention also provide a kind of meter of non-volatile program code that can be performed with processor
Calculation machine readable medium, said program code make the processor execute method described in first aspect.
The embodiment of the present invention brings following the utility model has the advantages that the embodiment of the present invention passes through first by original observed image data
It is divided into N number of rank according to image attributes, obtains classification image document corresponding to current time each rank, then using default
Deep neural network extended based on the classification image document of each rank, it is right to obtain subsequent time each rank institute
The extension data answered finally can carry out image synthesis to extension data corresponding to N number of rank, reconstruct prolonging for subsequent time
Stretch observed image data.
The embodiment of the present invention can be prolonged using deep neural network based on the classification image document of each rank
It stretches, obtains extension data corresponding to each rank of subsequent time, since the deep neural network is with preset strategy nerve
For the weight coefficient of network output as input, the weight coefficient can be such that the network state of the deep neural network is worth most
Greatly, the embodiment of the present invention can utilize the weight coefficient configurable deep nerve for the network state Maximum Value for making deep neural network
Network, and extend to obtain the extension data under residual error minimum using with the deep neural network postponed, it is based ultimately upon extension
Data reconstructs the extension observed image data of subsequent time, improves and extends accuracy.
Other features and advantages of the present invention will illustrate in the following description, also, partly become from specification
It obtains it is clear that understand through the implementation of the invention.The objectives and other advantages of the invention are in specification, claims
And specifically noted structure is achieved and obtained in attached drawing.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art be briefly described, it should be apparent that, it is described below
Attached drawing is some embodiments of the present invention, for those of ordinary skill in the art, before not making the creative labor
It puts, is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of flow chart of data extension method provided in an embodiment of the present invention；
Fig. 2 is the flow chart of step S101 in Fig. 1；
Fig. 3 is the schematic illustration that data provided in an embodiment of the present invention extends；
Fig. 4 is a kind of structure chart of data extension apparatus provided in an embodiment of the present invention.
Icon: 11- diversity module；12- prediction module；13- synthesis module.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Technical solution be clearly and completely described, it is clear that described embodiments are some of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, shall fall within the protection scope of the present invention.
Forecast system error is larger at present, is based on this, a kind of data extension method provided in an embodiment of the present invention, device and
Electronic equipment can use deep neural network and be extended based on the classification image document of each rank, obtained next
Extension data corresponding to moment each rank, since the deep neural network is with the power of preset tactful neural network output
For weight coefficient as input, the weight coefficient can make the network state Maximum Value of the deep neural network, and the present invention is real
Applying example can be using the weight coefficient configurable deep neural network for the network state Maximum Value for making deep neural network, and utilizes
Extend to obtain the extension data under residual error minimum with the deep neural network postponed, is based ultimately upon extension data and reconstructs down
The extension observed image data at one moment improves and extends accuracy.
For convenient for understanding the present embodiment, first to a kind of data extension method disclosed in the embodiment of the present invention into
Row is discussed in detail, as shown in Figure 1, the data extension method may comprise steps of:
Original observed image data is divided into N number of rank according to image attributes, obtains current time each grade by step S101
Not corresponding classification image document, wherein N is more than or equal to 1；
In embodiments of the present invention, original observed image data may include: that satellite data, Radar Data and automatic Weather Station are seen
Survey data etc..Image attributes can be down to image resolution ratio or graphical rule etc..
In embodiments of the present invention, as shown in Fig. 2, step S101 may comprise steps of:
Step S201 obtains original observed image data；
The original observed image data is carried out Wavelet image decomposition, obtains current time each rank by step S202
Corresponding classification image document.That is, the classification image document of N number of rank is obtained, the classification image document pair of different stage
Answer different image resolution ratio or graphical rule.
Wherein, Wavelet image decomposition method can provide the small echo classification image of different resolution or different scale, will divide
The small echo of different resolution or different scale after solution is classified image, the different stage image after decomposing as Wavelet image.
Step S102 is prolonged using preset deep neural network based on the classification image document of each rank
It stretches, obtains extension data corresponding to each rank of subsequent time.
In embodiments of the present invention, the deep neural network is made with the weight coefficient of preset tactful neural network output
For input, the weight coefficient makes the network state Maximum Value of the deep neural network；Deep neural network is for saving
The state estimation of the classification image document of each rank.Deep neural network establishes classification by a deep neural network
One neural network of (a certain rank) image is divided by certain strategy come the weight coefficient of percentage regulation neural network
The best estimate of grade (a certain rank) image, the clear image as the classification (a certain rank) under a certain resolution ratio.Depth
Neural network describes the state of atmosphere using full connection type, and weight is adjusted by weight and bias, and extension obtains a certain
Image under rank.
The strategy neural network includes: value assessment network and control neural network, and the value assessment network is used for
The value of the network state of the deep neural network is assessed, the control neural network is exported based on the value assessment network
Value and default adjustable strategies adjust the weight coefficient.
Tactful neural network is depth CNN neural network model, provides the adjustable strategies of valence deep neural network, provides depth
The value estimations value for spending neural network adjustment evaluates the value of deep neural network adjustment.Tactful neural network is main
It is that the weight coefficient of percentage regulation neural network is gone by the value of value assessment network query function.It is when being worth increase, then positive
The system of percentage regulation neural network moves the value of the value assessment network of the extension value of deep neural network toward direction is increased
It is dynamic.If value reduces, the weight coefficient of reversed percentage regulation neural network makes the value of the extension value of deep neural network
The value for assessing network is mobile toward opposite direction.Control neural network removes the weight coefficient of adjustment neural network, the depth nerve made
The value of network develops toward incremental direction, when the Maximum Value of value assessment network, then terminates network.
The default adjustable strategies of the control neural network are that RNN clustered control network is sent, the RNN clustered control
The value that network is used to export based on the value assessment network is to the control neural network output adjustment strategy.
RNN clustered control network provides a series of precision dimension parameter, provides adjustable strategies to tactful neural network
Policing parameter, RNN clustered control network model use cluster depth RNN neural network model, and there is forward and reverse to adjust plan
Slightly, the value estimations value effectively provided according to tactful neural network model, the tune of controlling depth neural network weight and bias
Perfect square to.
As shown in figure 3, the deep neural network includes multiple CNN neural networks, each CNN neural network and one
The classification image document of a rank is corresponding, is extended for the classification image document based on N grades；The strategy neural network
It is corresponded with the CNN neural network, the strategy neural network and the RNN clustered control network correspond.
This extension process can regard that the classification image document to each rank carries out extension extension respectively as, obtain
Subsequent time extension data at different levels.
Step S103 carries out image synthesis to extension data corresponding to N number of rank, and the extension for reconstructing subsequent time is seen
Altimetric image data.
In this step, can use the inverse transformations such as wavelet inverse transformation mode to extension data corresponding to N number of rank into
The synthesis of row image, reconstructs the extension observed image data of subsequent time, the extension observed image data of subsequent time can be with structure
At a clear image.
In this step, the value adjustment institute that the tactful neural network is used to export based on preset value assessment network
Weight coefficient is stated, specifically: when the value increases, forward direction adjusts the weight coefficient of the deep neural network, makes described
The value of the network state of deep neural network increases；If the value reduces, the weight of the neural network is reversely adjusted
Coefficient reduces the value of the network state of the deep neural network.
The embodiment of the present invention is worked as by the way that original observed image data is divided into N number of rank according to image attributes first
Then classification image document corresponding to preceding moment each rank utilizes institute of the preset deep neural network based on each rank
It states classification image document to be extended, obtains extension data corresponding to each rank of subsequent time, it finally can be to N number of rank
Corresponding extension data carries out image synthesis, reconstructs the extension observed image data of subsequent time.
The embodiment of the present invention can be prolonged using deep neural network based on the classification image document of each rank
It stretches, obtains extension data corresponding to each rank of subsequent time, since the deep neural network is with preset strategy nerve
For the weight coefficient of network output as input, the weight coefficient can be such that the network state of the deep neural network is worth most
Greatly, the embodiment of the present invention can utilize the weight coefficient configurable deep nerve for the network state Maximum Value for making deep neural network
Network, and extend to obtain the extension data under residual error minimum using with the deep neural network postponed, it is based ultimately upon extension
Data reconstructs the extension observed image data of subsequent time, improves and extends accuracy.
In another embodiment of the present invention, as shown in figure 4, also providing a kind of data extension apparatus, comprising:
Diversity module 11 obtains current time for original observed image data to be divided into N number of rank according to image attributes
Classification image document corresponding to each rank, wherein N is more than or equal to 1；
Extension of module 12, for using preset deep neural network based on the classification image document of each rank into
Row extends, and obtains extension data corresponding to each rank of subsequent time, the deep neural network is with preset strategy nerve
For the weight coefficient of network output as input, the weight coefficient makes the network state Maximum Value of the deep neural network；
Synthesis module 13 reconstructs subsequent time for carrying out image synthesis to extension data corresponding to N number of rank
Extend observed image data.
The technical effect and preceding method embodiment phase of device provided by the embodiment of the present invention, realization principle and generation
Together, to briefly describe, Installation practice part does not refer to place, can refer to corresponding contents in preceding method embodiment.
In another embodiment of the present invention, also offer a kind of electronic equipment, including memory, processor, the storage
The computer program that can be run on the processor is stored in device, the processor is realized when executing the computer program
The step of method described in above method embodiment.
In another embodiment of the present invention, a kind of non-volatile program code that can be performed with processor is also provided
Computer-readable medium, said program code make the processor execute method described in above method embodiment.
The flow chart and block diagram in the drawings show the system of multiple embodiments according to the present invention, method and computer journeys
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, section or code of table, a part of the module, section or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two continuous boxes can actually base
Originally it is performed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.It is also noted that
It is the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart, can uses and execute rule
The dedicated hardware based system of fixed function or movement is realized, or can use the group of specialized hardware and computer instruction
It closes to realize.
The computer program product of data extension method provided by the embodiment of the present invention, including storing program code
Computer readable storage medium, the instruction that said program code includes can be used for executing previous methods side as described in the examples
Method, specific implementation can be found in embodiment of the method, and details are not described herein.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description
It with the specific work process of device, can refer to corresponding processes in the foregoing method embodiment, details are not described herein.
In addition, in the description of the embodiment of the present invention unless specifically defined or limited otherwise, term " installation ", " phase
Even ", " connection " shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or be integrally connected；It can
To be mechanical connection, it is also possible to be electrically connected；It can be directly connected, can also can be indirectly connected through an intermediary
Connection inside two elements.For the ordinary skill in the art, above-mentioned term can be understood at this with concrete condition
Concrete meaning in invention.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
In the description of the present invention, it should be noted that term " center ", "upper", "lower", "left", "right", "vertical",
The orientation or positional relationship of the instructions such as "horizontal", "inner", "outside" be based on the orientation or positional relationship shown in the drawings, merely to
Convenient for description the present invention and simplify description, rather than the device or element of indication or suggestion meaning must have a particular orientation,
It is constructed and operated in a specific orientation, therefore is not considered as limiting the invention.In addition, term " first ", " second ",
" third " is used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance.
Finally, it should be noted that embodiment described above, only a specific embodiment of the invention, to illustrate the present invention
Technical solution, rather than its limitations, scope of protection of the present invention is not limited thereto, although with reference to the foregoing embodiments to this hair
It is bright to be described in detail, those skilled in the art should understand that: anyone skilled in the art
In the technical scope disclosed by the present invention, it can still modify to technical solution documented by previous embodiment or can be light
It is readily conceivable that variation or equivalent replacement of some of the technical features；And these modifications, variation or replacement, do not make
The essence of corresponding technical solution is detached from the spirit and scope of technical solution of the embodiment of the present invention, should all cover in protection of the invention
Within the scope of.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. a kind of data extension method characterized by comprising
Original observed image data is divided into N number of rank according to image attributes, obtains dividing corresponding to current time each rank
Grade image document, wherein N is more than or equal to 1；
Using preset deep neural network using the classification image document of each rank as input carry out data extension, obtain
Weight coefficient to extension data corresponding to each rank of subsequent time, the deep neural network uses preset strategy mind
The weight coefficient exported through network, the weight coefficient make the network state Maximum Value of the deep neural network；
Image synthesis is carried out to extension data corresponding to N number of rank, reconstructs the extension observed image data of subsequent time.
2. data extension method according to claim 1, which is characterized in that the strategy neural network includes: that value is commented
Estimate network and control neural network, the value assessment network is used to assess the valence of the network state of the deep neural network
Value, the value and the default adjustable strategies adjustment weight system that the control neural network export based on the value assessment network
Number.
3. data extension method according to claim 2, which is characterized in that the strategy neural network is used for based on default
Value assessment network output value adjustment described in weight coefficient, comprising:
When the value increases, forward direction adjusts the weight coefficient of the deep neural network, makes the deep neural network
The value of network state increases；
If the value reduces, the weight coefficient of the neural network is reversely adjusted, the network of the deep neural network is made
The value of state reduces.
4. data extension method according to claim 2, which is characterized in that the default adjustment plan of the control neural network
It is slightly that RNN clustered control network is sent, the RNN clustered control network is used for the valence exported based on the value assessment network
It is worth to the control neural network output adjustment strategy.
5. data extension method according to claim 4, which is characterized in that the deep neural network includes multiple CNN
Neural network, each CNN neural network is corresponding with the classification image document of a rank, for the classification based on N grades
Image document is extended；It is described strategy neural network and the CNN neural network correspond, it is described strategy neural network with
The RNN clustered control network corresponds.
6. data extension method according to claim 1, which is characterized in that it is described by original observed image data according to figure
As attribute is divided into N number of rank, classification image document corresponding to current time each rank is obtained, comprising:
Obtain original observed image data；
The original observed image data is subjected to Wavelet image decomposition, obtains classification figure corresponding to current time each rank
As data.
7. data extension method according to claim 1, which is characterized in that described to be provided to extension corresponding to N number of rank
Material carries out image synthesis, reconstructs the extension observed image data of subsequent time, comprising:
Image synthesis is carried out to extension data corresponding to N number of rank using wavelet inverse transformation, reconstructs the extension of subsequent time
Observed image data.
8. a kind of data extension apparatus characterized by comprising
Diversity module obtains current time each grade for original observed image data to be divided into N number of rank according to image attributes
Not corresponding classification image document, wherein N is more than or equal to 1；
Extension of module, for being prolonged using preset deep neural network based on the classification image document of each rank
It stretches, obtains extension data corresponding to each rank of subsequent time, the deep neural network is with preset tactful neural network
The weight coefficient of output makes the network state Maximum Value of the deep neural network as input, the weight coefficient；
Synthesis module, for carrying out image synthesis to extension data corresponding to N number of rank, the extension for reconstructing subsequent time is seen
Altimetric image data.
9. a kind of electronic equipment, including memory, processor, be stored in the memory to run on the processor
Computer program, which is characterized in that the processor realizes that the claims 1 to 7 are any when executing the computer program
The step of method described in item.
10. a kind of computer-readable medium for the non-volatile program code that can be performed with processor, which is characterized in that described
Program code makes the processor execute described any the method for claim 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811453086.8A CN109583568A (en) | 2018-11-28 | 2018-11-28 | Data extension method, device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811453086.8A CN109583568A (en) | 2018-11-28 | 2018-11-28 | Data extension method, device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109583568A true CN109583568A (en) | 2019-04-05 |
Family
ID=65923801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811453086.8A Pending CN109583568A (en) | 2018-11-28 | 2018-11-28 | Data extension method, device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109583568A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110187074A (en) * | 2019-06-12 | 2019-08-30 | 哈尔滨工业大学 | A kind of cross-country skiing racing track quality of snow prediction technique |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103926932A (en) * | 2014-04-25 | 2014-07-16 | 哈尔滨工程大学 | Intelligent ship moving posture decomposition field forecasting method |
| CN105976020A (en) * | 2016-04-28 | 2016-09-28 | 华北电力大学 | Network flow prediction method considering wavelet cross-layer correlations |
| CN106371155A (en) * | 2016-08-25 | 2017-02-01 | 华南师范大学 | A weather forecast method and system based on big data and analysis fields |
| CN106960253A (en) * | 2017-03-10 | 2017-07-18 | 国网浙江省电力公司经济技术研究院 | A kind of regional middle-term electricity demand forecasting method for considering great social activities influence |
| CN108182492A (en) * | 2017-12-29 | 2018-06-19 | 中科赛诺（北京）科技有限公司 | A kind of Data Assimilation method and device |
| CN108399248A (en) * | 2018-03-02 | 2018-08-14 | 郑州云海信息技术有限公司 | A kind of time series data prediction technique, device and equipment |
| CN108763865A (en) * | 2018-05-21 | 2018-11-06 | 成都信息工程大学 | A kind of integrated learning approach of prediction DNA protein binding sites |
| US20180329883A1 (en) * | 2017-05-15 | 2018-11-15 | Thomson Reuters Global Resources Unlimited Company | Neural paraphrase generator |
-
2018
- 2018-11-28 CN CN201811453086.8A patent/CN109583568A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103926932A (en) * | 2014-04-25 | 2014-07-16 | 哈尔滨工程大学 | Intelligent ship moving posture decomposition field forecasting method |
| CN105976020A (en) * | 2016-04-28 | 2016-09-28 | 华北电力大学 | Network flow prediction method considering wavelet cross-layer correlations |
| CN106371155A (en) * | 2016-08-25 | 2017-02-01 | 华南师范大学 | A weather forecast method and system based on big data and analysis fields |
| CN106960253A (en) * | 2017-03-10 | 2017-07-18 | 国网浙江省电力公司经济技术研究院 | A kind of regional middle-term electricity demand forecasting method for considering great social activities influence |
| US20180329883A1 (en) * | 2017-05-15 | 2018-11-15 | Thomson Reuters Global Resources Unlimited Company | Neural paraphrase generator |
| CN108182492A (en) * | 2017-12-29 | 2018-06-19 | 中科赛诺（北京）科技有限公司 | A kind of Data Assimilation method and device |
| CN108399248A (en) * | 2018-03-02 | 2018-08-14 | 郑州云海信息技术有限公司 | A kind of time series data prediction technique, device and equipment |
| CN108763865A (en) * | 2018-05-21 | 2018-11-06 | 成都信息工程大学 | A kind of integrated learning approach of prediction DNA protein binding sites |
Non-Patent Citations (5)
| Title |
|---|
| FRANCESCO CARLO MORABITO等: "Wavelet neural network processing of urban air pollution", 《PROCEEDINGS OF THE 2002 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS》 * |
| 丁红等: "小波变换集遗传算法神经网络的径流预测建模", 《广西大学学报:自然科学版》 * |
| 宋人杰等: "基于小波系数和BP神经网络的电力系统短期负荷预测", 《电力系统保护与控制》 * |
| 王红瑞等: "基于小波变换的支持向量机水文过程预测", 《清华大学学报(自然科学版)》 * |
| 高志强等编著: "《深度学习 从入门到实践》", 30 June 2018, 北京：中国铁道出版社 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110187074A (en) * | 2019-06-12 | 2019-08-30 | 哈尔滨工业大学 | A kind of cross-country skiing racing track quality of snow prediction technique |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Low et al. | Adaptive multi-robot wide-area exploration and mapping | |
| DE102018202497A1 (en) | Technologies for optimized machine learning training | |
| US20150254554A1 (en) | Information processing device and learning method | |
| CA3022125A1 (en) | System and method for improved neural network training | |
| Hu et al. | A dynamic closed-loop vehicle routing problem with uncertainty and incompatible goods | |
| CN86101057A (en) | Be used for method of allocating resources and equipment effectively | |
| US20200042901A1 (en) | Black-box optimization using neural networks | |
| CN108647810A (en) | The distribution method and device of order shipment, computer-readable medium | |
| CN108710109A (en) | A kind of trailer-mounted radar frequency band allocation method and system | |
| Alamir | A framework for real-time implementation of low-dimensional parameterized NMPC | |
| Chen et al. | Optimal structural policies for ambiguity and risk averse inventory and pricing models | |
| Pires et al. | Recommendation system using reinforcement learning for what-if simulation in digital twin | |
| CN109583568A (en) | Data extension method, device and electronic equipment | |
| CN106503386A (en) | The good and bad method and device of assessment luminous power prediction algorithm performance | |
| Wendong et al. | A multi-factor analysis model of quantitative investment based on GA and SVM | |
| CN108877276A (en) | Quick predict bus travel time method, apparatus and terminal | |
| Shang et al. | Immune clonal selection algorithm for capacitated arc routing problem | |
| Snoek et al. | Opportunity cost in Bayesian optimization | |
| Neydorf et al. | Improved bi-optimal hybrid approximation algorithm for monochrome multitone image processing | |
| Ma | A cross entropy multiagent learning algorithm for solving vehicle routing problems with time windows | |
| CN110378513A (en) | A kind of optimal solution for project planning determines method and system, equipment, medium | |
| JP6795240B1 (en) | Controls, methods and programs | |
| DE112020006904T5 (en) | Inference device, update method and update program | |
| WO2020023483A1 (en) | Continuous parametrizations of neural network layer weights | |
| Smolensky | Overview: Computational, Dynamical, and Statistical Perspectives on the Processing and Learning Problems in Neural Network Theory |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190405 |
|
| RJ01 | Rejection of invention patent application after publication |