CN106502799A - A kind of host load prediction method based on long memory network in short-term - Google Patents
A kind of host load prediction method based on long memory network in short-term Download PDFInfo
- Publication number
- CN106502799A CN106502799A CN201611271570.XA CN201611271570A CN106502799A CN 106502799 A CN106502799 A CN 106502799A CN 201611271570 A CN201611271570 A CN 201611271570A CN 106502799 A CN106502799 A CN 106502799A
- Authority
- CN
- China
- Prior art keywords
- load
- network
- term
- short
- prediction
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- 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
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
Abstract
The present invention is a kind of using long load predicting method of the memory network to cloud computing center main frame in short-term, belongs to cloud computing and deep learning field.The present invention solve problem be：In cloud computing environment, load on host computers changes violent problem.The present invention proposes the method that a kind of following load on host computers of prediction has improved cloud computing system scheduling.The core of the main algorithm of the present invention is to be modeled the relation between historical data and Future Data using the characteristic of long memory network in short-term, and there is network longterm memory function, output to be the load of prediction.The present invention utilizes the prediction loaded by the method for neutral net, by, compared with existing certain methods at present, method proposed by the present invention can be obtained and more accurately be predicted the outcome.
Description
Technical field
The invention belongs to cloud computing and deep learning field, mainly for load on host computers change in cloud computing environment acutely,
A kind of the features such as noise is big, it is proposed that new method based on the long host load prediction of memory network in short-term.
Background technology
In cloud computation data center, the change of load is generally very fierce, in order to tackle various complex situations in time, I
Need to be predicted the situation of various resources in data center.In cloud computation data center, the service condition of cpu resource is anti-
The ruuning situation of application program on main frame is answered, therefore when scheduling virtual machine is carried out, cpu resource is the resource of overriding concern.
When the load of certain main frame exceedes threshold value, the performance for running virtual machine on the host can be severely impacted, and therefore may be used
With by some virtual machine (vm) migrations on this main frame to other free hosts, so as to mitigate the load of main frame；When on some main frames
Load be less than threshold value when, the virtual machine on these main frames can be merged into other main frames or closing, so as to reduce cloud computing
The energy consumption at center.
Loading condition of some load predicting methods that presently, there are mainly for main frame in grid (Grid) calculating, but
In cloud computing environment, the type of main frame is different, and running on main frame for task is also different, and this is resulted in cloud environment
Load on host computers situation of change is more complicated, and certain methods before can not be entered to the loading condition of main frame in cloud environment well
Row prediction.
In order to solve problems of the prior art, this paper presents a kind of based on long memory network (Long in short-term
Short-Term Memory) method the context of load data is modeled, output is and final predicts the outcome.
Content of the invention
The purpose of the present invention：For the management that various types of main frames are carried out present cloud computing center centralization, unified
It is supplied to various types of users to use, causes under cloud computing environment, running on each main frame for task is different, which is born
The more complicated situation of the situation of change of load, it is proposed that a kind of can accurately predict following a period of time in load situation of change
Method, is the scheduling virtual machine of next step, and resource integrated management is laid a good foundation.
For problems of the prior art, the present invention proposes a kind of new host load prediction method, Neng Gouyou
Imitated is predicted to the load on host computers in following a period of time, and whole system is made up of following two main modulars：
Module one, the collection cloud computing load on host computers historical data of month, very long load sequence are divided into a lot of solid
Sizing, continuous historical data and prediction data, select suitable model to be modeled data.
Module two, make using training set long memory network in short-term learn corresponding parameter, and selected by cross validation collection
The best model of generalization is selected, output is final predicting the outcome.
For module one, related One-dimension Time Series, Ke Yitong before and after in the middle of cloud computing, the load of main frame is substantially
The method of a lot of time series analyses is crossed the prediction that loaded.In this module, we pass through to fix greatly load partition
Little subsequence is modeled so as to network.The input of network moment t is the historical data x=(d of one group of loadt-1,
dt-2..., dt-n), wherein each x represent t before length for n load histories sequence data.Corresponding it is output asWherein o be length for length after t for the future load of m prediction data.Assume true
Data be y=(dt+1, dt+2..., dt+m), the problem that we want to solve is to find a history value between future value
Mapping f：
Intuitively, closer to the history value of current time t, more related to predictive value.But, distant with current time
Historic load be also possible to provide some useful information, the trend for such as changing etc., help that we are loaded is pre-
Survey.So we want to construct a model, in-plant load information can be utilized, can also utilize load letter at a distance
Breath.
f(x；N)=f (g1(xt-1, xt-2..., xt-k), g2(xt-(k+1)..., xt-n))
Recognition with Recurrent Neural Network (Recurrent Neural Networks) is based on this simple thought, and tradition
Unlike propagated forward network, Recognition with Recurrent Neural Network has an intermediateness layer by the mechanism that feeds back so that it is very suitable
Preferably doing time series modeling.In addition, in order to be modeled to longterm memory, we are by the intermediateness layer of Recognition with Recurrent Neural Network
Long mnemon in short-term is replaced with, the framework of whole model is as shown in Figure 1.
Module two carries out load estimation using long memory network in short-term.
Once being previously used in view of certain information, forget that the state in this Geju City may have for network very much
With.In addition, compared to manually determining when to delete old state, it is intended that network oneself can decide when to delete
Remove.In this case, long memory network in short-term is undoubtedly one and selects well.
The framework of long memory network in short-term is as illustrated in fig. 2, it is assumed that the input of network moment t is historic load xt, output
For ht(final prediction o is obtained through a full articulamentum againt).Context in order to enable the network to load sequence is entered
Row modeling, we define a temporary location stFor store-memory.Meanwhile, in order to be able to enable load estimation to get long-time
Information, we are using the long feature definitions of memory network in short-term three gate functions (input layer, intermediate layer, output layers)：
Carry out the transmission of control information.Wherein, σ is sigmoid nonlinear activation functions, the weight and partially of W, U, b for network
Move.Then, we obtain finally entering for each moment network, intermediate value, output on this basis：
it=σ (b+Uxt+Wht-1)
When output is calculated, we select ReLu nonlinear activation functions to carry out the training of accelerator nerve network.
Then for a certain group of training data (x, y), its corresponding cost function is defined as by we
Wherein N be prediction length, oiIt is predictive value, yiIt is actual value.
Description of the drawings
Illustrate and technical solution of the present invention is further understood for providing, and constitute a part for description, with
The enforcement of the present invention for explaining technical scheme, does not constitute the restriction to technical solution of the present invention together.Accompanying drawing
It is described as follows：
Fig. 1 is the Organization Chart of whole model.
Fig. 2 is the Organization Chart of long memory network in short-term.
Specific embodiment
Describe embodiments of the present invention below with reference to accompanying drawing in detail, whereby to the present invention how application technology means
Carry out solve problem, and reach technique effect realize that process can fully understand and implement according to this.Illustrate in the flow process of accompanying drawing
Step can be executed in the different computer systems of such as one group of computer executable instructions, and, although in flow charts
Logical order is shown, but in some cases, shown or described step can be executed with the order being different from herein.
The implementation procedure of algorithm is specifically described below
Step 1-1, data acquisition.The historical data of n days load on host computers before collection.Using the CPU detection instruments on main frame
Every 5 minutes obtain CPU load data, which is divided into according to a certain percentage Training Set (training dataset),
Validation Set (cross validation collection) and Test Set (test data set), and very long load sequence is divided into a lot of solid
Sizing, continuous historical data and prediction data.
Step 1-2, e-learning.The long study of memory network in short-term is made using training set to corresponding parameter, by most
Littleization cost function, the weight coefficient for obtaining and the value of bias term, and the best model of generalization is selected by cross validation collection.
Step 1-3, carry out load estimation.Using the parameter for obtaining network from the study of long short term memory networking, input data
Collection, output are load on host computers.
Those skilled in the art should be understood that the system structure and each step of the above-mentioned present invention can be with general
Realizing, they can concentrate on single computing device computing device, or be distributed in the net of multiple computing devices compositions
On network, alternatively, they can be realized with the executable program code of computing device, it is thus possible to be stored in depositing
Executed by computing device in storage device, or they are fabricated to each integrated circuit modules respectively, or by them
Multiple modules or step are fabricated to single integrated circuit module to realize.So, the present invention is not restricted to any specific hardware
Combine with software.
Although the embodiment shown or described by the present invention is as above, described content is only to facilitate understand this
The embodiment that invents and adopt, is not limited to the present invention.Technical staff in any the technical field of the invention,
Without departing from disclosed herein spirit and scope on the premise of, can implement formal and details on do any repairing
Change and change, but the scope of patent protection of the present invention, still must be defined by the scope of which is defined in the appended claims.
Claims (3)
1. a kind of method and system based on the long host load prediction of memory network in short-term, is characterized in that comprising following main step
Suddenly：
Step one, the collection cloud computing load on host computers historical data of month, which is divided into training set, intersection according to a certain percentage
Checking collection and test set.
Step 2, very long load sequence is divided into a lot of fixed sizes, continuous historical datas and prediction data.
Step 3, the long study of memory network in short-term is made using training set to corresponding parameter, and selected by cross validation collection
The best model of generalization.
, in short-term in memory network, output is final predicting the outcome for step 4, the length for succeeding in school the data feeding of test set.
2. two it is characterised by the step of claim 1 methods described, we pass through the subsequence of load partition fixed size
So that network is modeled.The input of network moment t is the historical data x=(d of one group of loadt-1, dt-2..., dt-n), its
In each x represent t before length for n load histories sequence data.Corresponding it is output as
Wherein o be length for length after t for the future load of m prediction data.Assume that real data are y=(dt+1,
dt+2..., dt+m), then for a certain group of training data (x, y), its corresponding cost function is defined as by weWherein N be prediction length, oiIt is predictive value, yiIt is actual value.
3. the step of claim 1 methods described three be characterised by our Selection utilization length in short-term memory network being loaded
The modeling of prediction.Long memory network in short-term is one kind of Recognition with Recurrent Neural Network, it is assumed that the input of network moment t is historic load
xt, it is output as ht(final prediction o is obtained through a full articulamentum againt).Before and after enabling the network to load sequence
Relation is modeled, and we define a temporary location stFor store-memory.Meanwhile, in order to be able to enable load estimation to get
Prolonged information, we utilize the long feature definitions of memory network in short-term three gate functions (input layer, intermediate layer, output
Layer)： To control
The transmission of information processed.Then, we obtain finally entering for each moment network, intermediate value, output on this basis：it=σ
(b+Uxt+Wht-1),When output is calculated, we select ReLu activation letters
Several training to accelerate network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611271570.XA CN106502799A (en) | 2016-12-30 | 2016-12-30 | A kind of host load prediction method based on long memory network in short-term |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611271570.XA CN106502799A (en) | 2016-12-30 | 2016-12-30 | A kind of host load prediction method based on long memory network in short-term |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106502799A true CN106502799A (en) | 2017-03-15 |
Family
ID=58344974
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611271570.XA Pending CN106502799A (en) | 2016-12-30 | 2016-12-30 | A kind of host load prediction method based on long memory network in short-term |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106502799A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107391230A (en) * | 2017-07-27 | 2017-11-24 | 郑州云海信息技术有限公司 | A kind of implementation method and device for determining virtual machine load |
| CN107977748A (en) * | 2017-12-05 | 2018-05-01 | 中国人民解放军国防科技大学 | Multivariable distorted time sequence prediction method |
| CN108037378A (en) * | 2017-10-26 | 2018-05-15 | 上海交通大学 | Running state of transformer Forecasting Methodology and system based on long memory network in short-term |
| CN108170529A (en) * | 2017-12-26 | 2018-06-15 | 北京工业大学 | A kind of cloud data center load predicting method based on shot and long term memory network |
| CN108319501A (en) * | 2017-12-26 | 2018-07-24 | 中山大学 | A kind of flexible resource supply method and system based on micro services gateway |
| CN109117269A (en) * | 2018-07-26 | 2019-01-01 | 郑州云海信息技术有限公司 | A kind of distributed system dispatching method of virtual machine, device and readable storage medium storing program for executing |
| CN109714395A (en) * | 2018-12-10 | 2019-05-03 | 平安科技（深圳）有限公司 | Cloud platform resource uses prediction technique and terminal device |
| CN109829631A (en) * | 2019-01-14 | 2019-05-31 | 北京中兴通网络科技股份有限公司 | A kind of business risk early warning analysis method and system based on memory network |
| CN110191015A (en) * | 2019-05-21 | 2019-08-30 | 中山大学 | Cloud service performance intelligent Forecasting and device based on CPI index |
| CN110543411A (en) * | 2018-05-29 | 2019-12-06 | 南京大学 | Variation test intensity demand prediction method in evolution environment |
| CN111491006A (en) * | 2020-03-03 | 2020-08-04 | 天津大学 | Load-aware cloud computing resource elastic distribution system and method |
| CN111787109A (en) * | 2020-07-02 | 2020-10-16 | 哈尔滨工程大学 | Data center load balancing method based on time series prediction |
| CN112527470A (en) * | 2020-05-27 | 2021-03-19 | 上海有孚智数云创数字科技有限公司 | Model training method and device for predicting performance index and readable storage medium |
| CN112990586A (en) * | 2021-03-22 | 2021-06-18 | 海南电网有限责任公司澄迈供电局 | Intelligent video monitoring method and system for distribution network operation |
| CN113169990A (en) * | 2018-11-30 | 2021-07-23 | 阿里巴巴集团控股有限公司 | Segmentation of deep learning inference with dynamic offload |
| CN113220466A (en) * | 2021-06-02 | 2021-08-06 | 神州数码***集成服务有限公司 | Cloud service load universal prediction method based on long-term and short-term memory model |
| CN113220450A (en) * | 2021-04-29 | 2021-08-06 | 南京邮电大学 | Load prediction method, resource scheduling method and device for cloud-side multi-data center |
| CN115865684A (en) * | 2023-02-03 | 2023-03-28 | 浪潮电子信息产业股份有限公司 | Host scheduling method, system, equipment and computer readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7254534B2 (en) * | 2002-07-17 | 2007-08-07 | Stmicroelectronics N.V. | Method and device for encoding wideband speech |
| CN103748993B (en) * | 2009-05-31 | 2011-02-16 | 北京理工大学 | A kind of host load prediction method based on multisequencing combination |
| CN103365727A (en) * | 2013-07-09 | 2013-10-23 | 南京大学 | Host load forecasting method in cloud computing environment |
| CN103530190A (en) * | 2013-10-14 | 2014-01-22 | 北京邮电大学 | Load forecasting method and device |
| CN103678004A (en) * | 2013-12-19 | 2014-03-26 | 南京大学 | Host load prediction method based on unsupervised feature learning |
| CN105550323A (en) * | 2015-12-15 | 2016-05-04 | 北京国电通网络技术有限公司 | Load balancing prediction method of distributed database, and predictive analyzer |
| CN106022527A (en) * | 2016-05-27 | 2016-10-12 | 河南明晰信息科技有限公司 | Trajectory prediction method and device based on map tiling and LSTM cyclic neural network |
-
2016
- 2016-12-30 CN CN201611271570.XA patent/CN106502799A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7254534B2 (en) * | 2002-07-17 | 2007-08-07 | Stmicroelectronics N.V. | Method and device for encoding wideband speech |
| CN103748993B (en) * | 2009-05-31 | 2011-02-16 | 北京理工大学 | A kind of host load prediction method based on multisequencing combination |
| CN103365727A (en) * | 2013-07-09 | 2013-10-23 | 南京大学 | Host load forecasting method in cloud computing environment |
| CN103530190A (en) * | 2013-10-14 | 2014-01-22 | 北京邮电大学 | Load forecasting method and device |
| CN103678004A (en) * | 2013-12-19 | 2014-03-26 | 南京大学 | Host load prediction method based on unsupervised feature learning |
| CN105550323A (en) * | 2015-12-15 | 2016-05-04 | 北京国电通网络技术有限公司 | Load balancing prediction method of distributed database, and predictive analyzer |
| CN106022527A (en) * | 2016-05-27 | 2016-10-12 | 河南明晰信息科技有限公司 | Trajectory prediction method and device based on map tiling and LSTM cyclic neural network |
Non-Patent Citations (3)
| Title |
|---|
| HOCHREITER S, SCHMIDHUBER J: "Long short-term memory", 《NEURAL COMPUT》 * |
| YOUNES SAMIH ET AL: "Multiilingual Code-switching Identification via LSTM Recurrent Neural Networks", 《PROCEEDINGS OF THE SECOND WORKSHOP ON COMPUTATIONAL APPROACHES TO CODE SWITCHING》 * |
| 李彦冬等: "卷积神经网络研究综述", 《计算机应用》 * |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107391230A (en) * | 2017-07-27 | 2017-11-24 | 郑州云海信息技术有限公司 | A kind of implementation method and device for determining virtual machine load |
| CN108037378B (en) * | 2017-10-26 | 2020-08-07 | 上海交通大学 | Transformer operation state prediction method and system based on long-time and short-time memory network |
| CN108037378A (en) * | 2017-10-26 | 2018-05-15 | 上海交通大学 | Running state of transformer Forecasting Methodology and system based on long memory network in short-term |
| CN107977748A (en) * | 2017-12-05 | 2018-05-01 | 中国人民解放军国防科技大学 | Multivariable distorted time sequence prediction method |
| CN107977748B (en) * | 2017-12-05 | 2022-03-11 | 中国人民解放军国防科技大学 | Multivariable distorted time sequence prediction method |
| CN108319501A (en) * | 2017-12-26 | 2018-07-24 | 中山大学 | A kind of flexible resource supply method and system based on micro services gateway |
| CN108319501B (en) * | 2017-12-26 | 2022-01-14 | 中山大学 | Elastic resource supply method and system based on micro-service gateway |
| CN108170529A (en) * | 2017-12-26 | 2018-06-15 | 北京工业大学 | A kind of cloud data center load predicting method based on shot and long term memory network |
| CN110543411A (en) * | 2018-05-29 | 2019-12-06 | 南京大学 | Variation test intensity demand prediction method in evolution environment |
| CN110543411B (en) * | 2018-05-29 | 2021-07-02 | 南京大学 | Variation test intensity demand prediction method in evolution environment |
| CN109117269A (en) * | 2018-07-26 | 2019-01-01 | 郑州云海信息技术有限公司 | A kind of distributed system dispatching method of virtual machine, device and readable storage medium storing program for executing |
| CN113169990B (en) * | 2018-11-30 | 2024-04-05 | 阿里巴巴集团控股有限公司 | Segmentation of deep learning reasoning with dynamic offloading |
| CN113169990A (en) * | 2018-11-30 | 2021-07-23 | 阿里巴巴集团控股有限公司 | Segmentation of deep learning inference with dynamic offload |
| CN109714395A (en) * | 2018-12-10 | 2019-05-03 | 平安科技（深圳）有限公司 | Cloud platform resource uses prediction technique and terminal device |
| CN109829631A (en) * | 2019-01-14 | 2019-05-31 | 北京中兴通网络科技股份有限公司 | A kind of business risk early warning analysis method and system based on memory network |
| CN110191015A (en) * | 2019-05-21 | 2019-08-30 | 中山大学 | Cloud service performance intelligent Forecasting and device based on CPI index |
| CN111491006A (en) * | 2020-03-03 | 2020-08-04 | 天津大学 | Load-aware cloud computing resource elastic distribution system and method |
| CN112527470A (en) * | 2020-05-27 | 2021-03-19 | 上海有孚智数云创数字科技有限公司 | Model training method and device for predicting performance index and readable storage medium |
| CN111787109A (en) * | 2020-07-02 | 2020-10-16 | 哈尔滨工程大学 | Data center load balancing method based on time series prediction |
| CN112990586A (en) * | 2021-03-22 | 2021-06-18 | 海南电网有限责任公司澄迈供电局 | Intelligent video monitoring method and system for distribution network operation |
| CN113220450A (en) * | 2021-04-29 | 2021-08-06 | 南京邮电大学 | Load prediction method, resource scheduling method and device for cloud-side multi-data center |
| CN113220450B (en) * | 2021-04-29 | 2022-10-21 | 南京邮电大学 | Load prediction method, resource scheduling method and device for cloud-side multi-data center |
| CN113220466A (en) * | 2021-06-02 | 2021-08-06 | 神州数码***集成服务有限公司 | Cloud service load universal prediction method based on long-term and short-term memory model |
| CN115865684A (en) * | 2023-02-03 | 2023-03-28 | 浪潮电子信息产业股份有限公司 | Host scheduling method, system, equipment and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106502799A (en) | A kind of host load prediction method based on long memory network in short-term | |
| US11100266B2 (en) | Generating integrated circuit floorplans using neural networks | |
| Wu et al. | An evolutionary fuzzy scheduler for multi-objective resource allocation in fog computing | |
| CN104636801A (en) | Transmission line audible noise prediction method based on BP neural network optimization | |
| CN102694800A (en) | Gaussian process regression method for predicting network security situation | |
| CN106789214A (en) | It is a kind of based on the just remaining pair network situation awareness method and device of string algorithm | |
| Shi et al. | Artificial bee colony Based Bayesian Regularization Artificial Neural Network approach to model transient flammable cloud dispersion in congested area | |
| US20190228297A1 (en) | Artificial Intelligence Modelling Engine | |
| CN115130711A (en) | Data processing method and device, computer and readable storage medium | |
| CN109583659A (en) | User's operation behavior prediction method and system based on deep learning | |
| Said et al. | AI-based solar energy forecasting for smart grid integration | |
| Battiti et al. | Reactive search optimization: learning while optimizing | |
| Peña et al. | A novel multi-objective optimization approach to guarantee quality of service and energy efficiency in a heterogeneous bus fleet system | |
| Zhao et al. | Adaptive Swarm Intelligent Offloading Based on Digital Twin-assisted Prediction in VEC | |
| CN116595356B (en) | Time sequence signal prediction method and device, electronic equipment and storage medium | |
| Tahk et al. | A hybrid optimization method of evolutionary and gradient search | |
| CN113610299B (en) | Information propagation prediction method and device based on characteristic attenuation reinforced neural network | |
| CN115587616A (en) | Network model training method and device, storage medium and computer equipment | |
| Hussain et al. | Forecasting time series stock data using deep learning technique in a distributed computing environment | |
| CN113344369A (en) | Method and device for attributing image data, electronic equipment and storage medium | |
| Yang et al. | Energy saving strategy of cloud data computing based on convolutional neural network and policy gradient algorithm | |
| Samanta et al. | Energy management in hybrid electric vehicles using optimized radial basis function neural network | |
| Xiang et al. | HDFS efficiency storage strategy for big data in smart city | |
| Wang et al. | Charging load forecasting of electric vehicles based on sparrow search algorithm‐improved random forest regression model | |
| Lin et al. | Thinking and methodology of multi-objective optimization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170315 |
|
| WD01 | Invention patent application deemed withdrawn after publication |