CN114584476A - Traffic prediction method, network training device and electronic equipment - Google Patents
Traffic prediction method, network training device and electronic equipment Download PDFInfo
- Publication number
- CN114584476A CN114584476A CN202011284570.XA CN202011284570A CN114584476A CN 114584476 A CN114584476 A CN 114584476A CN 202011284570 A CN202011284570 A CN 202011284570A CN 114584476 A CN114584476 A CN 114584476A
- Authority
- CN
- China
- Prior art keywords
- data
- prediction
- time point
- network
- adjustment
- 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
- 238000000034 method Methods 0.000 title claims abstract description 87
- 238000012549 training Methods 0.000 title claims abstract description 68
- 238000012545 processing Methods 0.000 claims description 75
- 230000015654 memory Effects 0.000 claims description 31
- 238000012937 correction Methods 0.000 claims description 22
- 238000004590 computer program Methods 0.000 claims description 11
- 239000000126 substance Substances 0.000 claims description 6
- 230000000737 periodic effect Effects 0.000 description 18
- 238000010586 diagram Methods 0.000 description 8
- 230000001360 synchronised effect Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000013468 resource allocation Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 235000019800 disodium phosphate Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Environmental & Geological Engineering (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The embodiment of the invention discloses a flow prediction method, a network training method, a device and electronic equipment. The network training method comprises the following steps: obtaining a data set containing a plurality of first data, and obtaining first prediction data of a prediction time point based on the data set and a first network; the plurality of first data are historical flow data before the predicted time point; obtaining a plurality of second data, and obtaining adjustment data based on the plurality of second data and a second network; adjusting the first prediction data according to the adjustment data and the dynamic adjustment parameters to obtain second prediction data, and generating prediction sequence data based on the obtained plurality of second prediction data; and after the prediction sequence data is processed by a fully-connected network, determining loss with target sequence data, and training the first network, the second network and the fully-connected network according to the loss.
Description
Technical Field
The invention relates to the field of traffic prediction, in particular to a traffic prediction method, a network training device and electronic equipment.
Background
In recent years, the usage amount of internet traffic of users is increased year by year, and particularly, the gradual application of the 5G network in the world is more demanding for operators. An operator needs to deal with the problem of network resource allocation to improve service efficiency, and when certain node resources are not allocated sufficiently, supersaturation of users is caused, node service load is increased, and user experience is reduced; on the contrary, when the node resource allocation is too much, the resource waste is caused, and even the service pressure of the peripheral nodes is too high. In recent years, some common deep learning methods are also applied to the application of traffic prediction.
The current flow prediction technology does not consider the influence of periodic data on the prediction, or the considered periodic data is not strict periodic data for the flow prediction, and the flow may have certain dynamic fluctuation at the time point of the same period, so that the prediction result has certain deviation.
Disclosure of Invention
In order to solve the existing technical problems, embodiments of the present invention provide a traffic prediction method, a network training method, an apparatus, and an electronic device.
In order to achieve the above purpose, the technical solution of the embodiment of the present invention is realized as follows:
in a first aspect, an embodiment of the present invention provides a network training method for traffic prediction, where the method includes:
obtaining a data set containing a plurality of first data, and obtaining first prediction data of a prediction time point based on the data set and a first network; the plurality of first data are historical flow data before the predicted time point;
obtaining a plurality of second data, and obtaining adjustment data based on the plurality of second data and a second network, wherein the plurality of second data are historical flow data in the same period as the predicted time point; the plurality of second data comprise historical flow data of time points which are in the same period as the predicted time point and also comprise historical flow data of time points before and/or after the time points which are in the same period as the predicted time point;
adjusting the first prediction data according to the adjustment data and the dynamic adjustment parameters to obtain second prediction data, and generating prediction sequence data based on the obtained plurality of second prediction data;
and after the predicted sequence data is processed by a fully-connected network, determining loss with target sequence data, and training the first network, the second network and the fully-connected network according to the loss.
In the above scheme, the method further comprises: adding the second prediction data to the data set, and deleting one first data in the data set to obtain an updated data set;
retrieving first predicted data for the predicted point in time based on the updated data set and the first network.
In the above solution, the plurality of second data includes first historical flow data at a time point in the same period as the predicted time point, second historical flow data at a time point before the time point in the same period as the predicted time point, and third historical flow data at a time point after the time point in the same period as the predicted time point;
obtaining adjustment data based on the plurality of second data and the second network includes:
respectively obtaining first adjustment data, second adjustment data and third adjustment data corresponding to the first historical traffic data, the second historical traffic data and the third historical traffic data based on the second network;
and performing weighting processing on the first adjustment data, the second adjustment data and the third adjustment data, and obtaining the adjustment data according to a weighting processing result.
In the foregoing solution, the performing weighting processing on the first adjustment data, the second adjustment data, and the third adjustment data includes:
and taking a dynamic adjustment parameter alpha as a weight of the first adjustment data, and taking (1-alpha)/2 as weights of the second adjustment data and the third adjustment data, and performing weighted summation processing on the first adjustment data, the second adjustment data and the third adjustment data to obtain a weighted processing result.
In the foregoing solution, the adjusting the first prediction data according to the adjustment data and the dynamic adjustment parameter to obtain second prediction data includes:
and obtaining a correction value by utilizing the product of the dynamic correction factor beta and the adjustment data, and adding the correction value and the first prediction data to obtain the second prediction data.
In the above solution, the determining the loss with the target sequence data after the prediction sequence data is processed by the fully-connected network includes:
obtaining third data in the prediction sequence data processed by the full-connection network, and obtaining fourth data in the target sequence data; the third data is the ith data in the predicted sequence data, and the fourth data is the ith data in the target sequence data; the value of i is greater than 0 and less than or equal to the number of data contained in the prediction sequence data or the target sequence data;
determining a mean absolute percentage error between the third data and the fourth data, and determining a cosine distance between the third data and the fourth data, determining a loss based on the mean absolute percentage error and the cosine distance.
In a second aspect, an embodiment of the present invention further provides a traffic prediction method, where the method includes:
obtaining first class data and second class data, wherein the first class data comprises a plurality of historical flow data before a predicted time point, the second class data comprises a plurality of historical flow data in the same period as the predicted time point, the plurality of second class data comprises the historical flow data of the time point in the same period as the predicted time point, and the historical flow data of the time point before and/or after the time point in the same period as the predicted time point;
obtaining initial prediction data based on the first type of data and a first network obtained by training, and obtaining adjustment data based on the second type of data and a second network obtained by training;
adjusting the initial prediction data according to the adjustment data and the dynamic adjustment parameters to obtain prediction data;
and generating prediction sequence data based on the plurality of obtained prediction data, and processing the prediction sequence data through a full-connection network to obtain a flow prediction result at the prediction time point.
In the above solution, the first network, the second network and the fully connected network are obtained by training based on the method of the foregoing first aspect of the embodiment of the present invention.
In a third aspect, an embodiment of the present invention further provides a network training apparatus for traffic prediction, where the apparatus includes: the device comprises a first processing unit, a second processing unit, an adjusting unit and a training unit; wherein the content of the first and second substances,
the first processing unit is used for obtaining a data set containing a plurality of first data and obtaining first prediction data of a prediction time point based on the data set and a first network; the plurality of first data are historical flow data before the predicted time point;
the second processing unit is configured to obtain a plurality of second data, and obtain adjustment data based on the plurality of second data and a second network, where the plurality of second data are historical traffic data in the same period as the predicted time point; the plurality of second data comprise historical flow data of time points which are in the same period as the predicted time point and also comprise historical flow data of time points before and/or after the time points which are in the same period as the predicted time point;
the adjusting unit is used for adjusting the first prediction data according to the adjustment data and the dynamic adjustment parameters to obtain second prediction data, and generating prediction sequence data based on the obtained second prediction data;
and the training unit is used for determining loss with target sequence data after the prediction sequence data is processed by a fully-connected network, and training the first network, the second network and the fully-connected network according to the loss.
In the foregoing solution, the first processing unit is further configured to add the second prediction data to the data set, and delete one first data in the data set to obtain an updated data set; retrieving first predicted data for the predicted point in time based on the updated data set and the first network.
In the above solution, the plurality of second data includes first historical flow data at a time point in the same period as the predicted time point, second historical flow data at a time point before the time point in the same period as the predicted time point, and third historical flow data at a time point after the time point in the same period as the predicted time point;
the second processing unit is configured to obtain first adjustment data, second adjustment data, and third adjustment data corresponding to the first historical traffic data, the second historical traffic data, and the third historical traffic data, respectively, based on the second network; and performing weighting processing on the first adjustment data, the second adjustment data and the third adjustment data, and obtaining the adjustment data according to a weighting processing result.
In the foregoing solution, the second processing unit is configured to perform weighted summation on the first adjustment data, the second adjustment data, and the third adjustment data by using a dynamic adjustment parameter α as a weight of the first adjustment data, and using (1- α)/2 as weights of the second adjustment data and the third adjustment data, so as to obtain a weighted processing result.
In the foregoing solution, the adjusting unit is configured to obtain a correction value by using a product of a dynamic correction factor β and the adjustment data, and add the correction value and the first prediction data to obtain the second prediction data.
In the above scheme, the training unit is configured to obtain third data in the predicted sequence data processed by the fully-connected network, and obtain fourth data in the target sequence data; the third data is the ith data in the predicted sequence data, and the fourth data is the ith data in the target sequence data; the value of i is greater than 0 and less than or equal to the number of data contained in the prediction sequence data or the target sequence data; determining a mean absolute percentage error between the third data and the fourth data, and determining a cosine distance between the third data and the fourth data, determining a loss based on the mean absolute percentage error and the cosine distance.
In a fourth aspect, an embodiment of the present invention further provides a flow prediction apparatus, where the apparatus includes: a data acquisition unit and a third processing unit; wherein the content of the first and second substances,
the data acquisition unit is used for acquiring first-class data and second-class data, wherein the first-class data comprises a plurality of historical flow data before a predicted time point, the second-class data comprises a plurality of historical flow data in the same period as the predicted time point, the plurality of second data comprises the historical flow data at the same period as the predicted time point and the historical flow data at time points before and/or after the same period as the predicted time point;
the third processing unit is used for obtaining initial prediction data based on the first type of data and a first network obtained by training; obtaining adjustment data based on the second type of data and a second network obtained through training; adjusting the initial prediction data according to the adjustment data and the dynamic adjustment parameters to obtain prediction data; and generating prediction sequence data based on the plurality of obtained prediction data, and processing the prediction sequence data through a full-connection network to obtain a flow prediction result at the prediction time point.
In the above solution, the first network, the second network and the fully-connected network are obtained by training based on the network training apparatus for traffic prediction according to the third aspect of the embodiment of the present invention.
In a fifth aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the method according to the foregoing first aspect or second aspect.
In a sixth aspect, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps of the method according to the foregoing first aspect or second aspect of the embodiment of the present invention.
The embodiment of the invention provides a traffic prediction method, a network training device and electronic equipment, wherein the network training method comprises the following steps: obtaining a data set containing a plurality of first data, and obtaining first prediction data of a prediction time point based on the data set and a first network; the plurality of first data are historical flow data before the predicted time point; obtaining a plurality of second data, and obtaining adjustment data based on the plurality of second data and a second network, wherein the plurality of second data are historical flow data in the same period as the predicted time point; the plurality of second data comprise historical flow data of time points which are in the same period as the predicted time point and also comprise historical flow data of time points before and/or after the time points which are in the same period as the predicted time point; adjusting the first prediction data according to the adjustment data and the dynamic adjustment parameters to obtain second prediction data, and generating prediction sequence data based on the obtained plurality of second prediction data; and after the prediction sequence data is processed by a fully-connected network, determining loss with target sequence data, and training the first network, the second network and the fully-connected network according to the loss. By adopting the technical scheme of the embodiment of the invention, on one hand, the dynamic time compactness characteristic of the historical flow data is learned through the first network, the periodic characteristic of the periodic data is learned through the second network, and the adjusting data is determined based on the historical flow data of the time point and the time points before and after the time point which have the same period as the predicting time point, so that the first predicting data is adjusted based on the adjusting data, namely, the influence of the periodic data is considered in the flow predicting process; on the other hand, in the flow prediction process, the influence of the periodic data and the data before and/or after the periodic time point on the prediction is considered, and the dynamic adjustment parameter is combined with the adjustment data to adjust the first prediction data, so that the accuracy of flow prediction is improved.
Drawings
Fig. 1 is a schematic flow chart of a network training method for traffic prediction according to an embodiment of the present invention;
FIG. 2 is a schematic block diagram of a processing flow of a network training method for traffic prediction according to an embodiment of the present invention;
FIG. 3 is a flow chart illustrating a flow prediction method according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a network training apparatus for traffic prediction according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a flow prediction apparatus according to an embodiment of the present invention;
fig. 6 is a schematic diagram of a hardware component structure of the electronic device according to the embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
The terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the disclosure. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
The embodiment of the invention provides a network training method for flow prediction. Embodiments of the present invention may be applied to various electronic devices including, but not limited to, fixed devices and/or mobile devices, for example, the fixed devices include, but are not limited to: a Personal Computer (PC), a server, which may be a cloud server or a general server, or the like.
Fig. 1 is a schematic flowchart of a network training method for traffic prediction according to an embodiment of the present invention; as shown in fig. 1, the method includes:
step 101: obtaining a data set containing a plurality of first data, and obtaining first prediction data of a prediction time point based on the data set and a first network; the plurality of first data are historical flow data before the predicted time point;
step 102: obtaining a plurality of second data, and obtaining adjustment data based on the plurality of second data and a second network, wherein the plurality of second data are historical flow data in the same period as the predicted time point; the plurality of second data comprise historical flow data of time points which are in the same period as the predicted time point and also comprise historical flow data of time points before and/or after the time points which are in the same period as the predicted time point;
step 103: adjusting the first prediction data according to the adjustment data to obtain second prediction data, and generating prediction sequence data based on the obtained second prediction data;
step 104: and after the prediction sequence data is processed by a fully-connected network, determining loss with target sequence data, and training the first network, the second network and the fully-connected network according to the loss.
In this embodiment, the first network is a deep learning network, and may be specifically a recurrent neural network. Illustratively, the first network may employ a Long Short-Term Memory (LSTM) network. In the embodiment, a plurality of historical traffic data are analyzed and processed through the first network, and the time compactness characteristic is learned, so that first prediction data of a prediction time point is obtained through prediction.
The plurality of first data included in the data set are specifically a plurality of data before the predicted time point, and for example, the plurality of first data may be a plurality of historical flow data satisfying a certain time parameter. For example, a plurality of historical traffic data may be randomly obtained within a time range of a preset time period before the predicted time point from the historical traffic data. Optionally, each piece of first data represents historical traffic information corresponding to a historical time or a historical time period, that is, the first data may carry a time parameter.
In this embodiment, the second network is used to learn periodic characteristics. Illustratively, the second network may be implemented by a Multi-Layer Perceptron (MLP) network. In the embodiment, the historical traffic data in the same period as the predicted time point is analyzed and processed through the second network, and the periodic characteristics are learned.
Each of the second data represents a historical flow rate corresponding to a historical time (time point). The plurality of second data at least include historical flow data of a time point in the same period as the preset time point, and may further include historical flow data of a time point before and/or a time point after the preset time point in the same period. Illustratively, the second data may be represented in the form of a sequence.
In view of that the disturbance caused by the periodic dynamic change is generally at a position where the time point adjacent to the predicted time point in the same period is short, in this embodiment, the plurality of second data includes historical flow data of the time point in the same period as the predicted time point, and also includes historical flow data of time points before and/or after the time point in the same period as the predicted time point. In some optional embodiments, the plurality of second data includes first historical flow data of a time point in the same period as the predicted time point, second historical flow data of a time point before the time point in the same period as the predicted time point, and third historical flow data of a time point after the time point in the same period as the predicted time point; the time point in the same period as the predicted time point may be selected as the first time point, the time points before and after the first time point may be respectively selected as the second time point and the third time point, and the first historical flow data corresponding to the first time point, the second historical flow data corresponding to the second time point, and the third historical flow data corresponding to the third time point may be respectively selected as the second data. Of course, the second data in this embodiment is not limited to the above example, and time points separated from the first time point by a preset time may also be used as the second time point and the third time point, and the length of the preset time may be set according to actual situations.
In this embodiment, the adjustment data obtained based on the plurality of second data and the second network is used to adjust the first prediction data, so as to adjust the accuracy of the single prediction data (i.e., the first prediction data).
In some optional embodiments of the invention, the plurality of second data includes first historical flow data of a time point in a same period as the predicted time point, second historical flow data of a time point before the time point in the same period as the predicted time point, and third historical flow data of a time point after the time point in the same period as the predicted time point; obtaining adjustment data based on the plurality of second data and the second network includes: respectively obtaining first adjustment data, second adjustment data and third adjustment data corresponding to the first historical traffic data, the second historical traffic data and the third historical traffic data based on the second network; and performing weighting processing on the first adjustment data, the second adjustment data and the third adjustment data, and obtaining the adjustment data according to a weighting processing result.
In some optional embodiments, the weighting the first adjustment data, the second adjustment data, and the third adjustment data includes: and taking a dynamic adjustment parameter alpha as a weight of the first adjustment data, and taking (1-alpha)/2 as weights of the second adjustment data and the third adjustment data, and performing weighted summation processing on the first adjustment data, the second adjustment data and the third adjustment data to obtain a weighted processing result.
In this embodiment, taking an example that the plurality of second data includes first historical flow data of a time point at which a predicted time point has the same period, second historical flow data of a time point before the predicted time point has the same period, and third historical flow data of a time point after the predicted time point has the same period, the first historical flow data, the second historical flow data, and the third flow data are respectively processed by the second network, so as to obtain first adjustment data (denoted as Q1) corresponding to the first historical flow data, second adjustment data (denoted as Q2) corresponding to the second historical flow data, and third adjustment parameters (denoted as Q2) corresponding to the third historical flow data, respectively, and taking a parameter α as a weight of the first adjustment data and taking (1- α)/2 as weights of the second adjustment data and the third adjustment data, the adjustment data Q satisfy:
wherein the value range of alpha is more than 0 and less than 1. Of course, the value range of α in the present embodiment is not limited to the above example.
In some optional embodiments of the present invention, the adjusting the first prediction data according to the adjustment data to obtain second prediction data includes: and obtaining a correction value by utilizing the product of the dynamic correction factor and the adjustment data, and adding the correction value and the first prediction data to obtain the second prediction data.
In this embodiment, if the first prediction data is set as S1, the adjustment data is still set as Q, and the correction factor is set as β, the second prediction data S2 satisfies:
S2＝S1+β×Q (2)
illustratively, β has a value in the range of greater than-0.5 to less than 0.5. Of course, the value range of β in the present embodiment is not limited to the above example.
In this embodiment, in some optional embodiments, a plurality of second prediction data may be obtained according to the processing of steps 101 to 102 described above based on the obtained plurality of sets of data.
In further alternative embodiments, the method further comprises: adding the second prediction data to the data set, and deleting one first data in the data set to obtain an updated data set; retrieving first predicted data for the predicted point in time based on the updated data set and the first network.
In this embodiment, the data set may be a data sequence arranged in a certain order, and after the first second prediction data is obtained, the second prediction data is added to the data set (i.e. the data sequence), and delete the first data in the data set, i.e. the data sequence, resulting in a new data set, further, according to the processing from the step 101 to the step 102, second prediction data is obtained, and then repeatedly adding the second prediction data to the data set (i.e. the data sequence) and deleting the first data in the data set (i.e. the data sequence), thereby obtaining a new data set, further, third second prediction data is obtained according to the processing from step 101 to step 102, and repeating the steps until prediction sequence data are obtained, wherein the prediction sequence data comprise a set number of second prediction data.
In some optional embodiments of the present invention, the determining the loss with the target sequence data after processing the prediction sequence data through the fully connected network comprises: obtaining third data in the predicted sequence data and fourth data in the target sequence data; the third data is the ith data in the predicted sequence data, and the fourth data is the ith data in the target sequence data; the value of i is greater than 0 and less than or equal to the number of data contained in the prediction sequence data or the target sequence data; determining a mean absolute percentage error between the third data and the fourth data, and determining a cosine distance between the third data and the fourth data, determining a loss based on the mean absolute percentage error and the cosine distance.
In this embodiment, the loss may adopt a mapping Cosine Distance (M-C, MAPE-Cosine Distance) mixed loss function, that is, the loss includes two parts: one part is the mean absolute percentage error between the predicted data and the actual data (i.e., the target data), and the other part is the cosine distance between the predicted data and the actual data (i.e., the target data).
Illustratively, the loss L satisfies:
wherein k is a hyper-parameter for adjusting cosine distance specific gravity, the value of k is less than-1, yiIs the ith value, p, of the target sequence dataiIs the ith value of the predicted sequence data; n represents the number of data in the target sequence data and the predicted sequence data.
In this embodiment, after the predicted sequence data obtained in step 103 is processed through the fully-connected network, the loss L is determined according to the formula (3) according to the processing result of the fully-connected network and the target sequence data, and parameters in the first network, the second network and the fully-connected network are adjusted according to the loss L, that is, the first network, the second network and the fully-connected network are trained, so that the loss L obtained based on the predicted sequence data and the target sequence data satisfies the convergence condition.
By adopting the technical scheme of the embodiment of the invention, on the first hand, the dynamic time compactness characteristic of the historical flow data is learned through the first network, the periodic characteristic of the periodic data is learned through the second network, and the adjusting data is determined based on the historical flow data of the time point and the time points before and after the time point which have the same period as the predicting time point, so that the first predicting data is adjusted based on the adjusting data, namely, the influence of the periodic data is considered in the flow predicting process; in the second aspect, in the process of predicting the flow, the influence of the periodic data and data before and/or after the periodic time point on the prediction is considered, and the dynamic adjustment parameters (including the dynamic adjustment parameter α and the dynamic correction factor β) are combined with the adjustment data to adjust the first prediction data, so that the accuracy of the single prediction data (for example, the second prediction data) is improved. In a third aspect, the loss function of the present application includes an average absolute percentage error and a cosine distance, which can reduce the deviation between the predicted data and the actual data, thereby improving the accuracy of the network and the accuracy of the traffic prediction.
The following describes a network training method according to an embodiment of the present invention with reference to a specific example.
FIG. 2 is a schematic block diagram of a processing flow of a network training method for traffic prediction according to an embodiment of the present invention; as shown in fig. 2, in the present embodiment, the input data includes: three periodic sequences of dimension n × P { P0}n×p、{P-1}n×p、{P1}n×pAnd a preamble time sequence of length W Ww×1，{W}w×1W data are included. Wherein, the above { W }w×1The data set in the previous embodiment is the above { P }0}n×p、{P-1}n×p、{P1}n×pThe second data in the foregoing embodiment. The output data is a prediction sequence (W) with the length of pp×1。
Step 2, a sequence { P) of corresponding time points and adjacent time points of n periods before the predicted time point0}n×i、{P-1}n×i、{P1}n×iRespectively pass through a second network and are weighted by a dynamic parameter alpha to output adjustment data S2i；
Wherein, suppose { P0}n×i、{P-1}n×i、{P1}n×iThe output results through the second network are respectively marked as Q1, Q2 and Q3, then the data S is adjusted2iSatisfies the following conditions:
wherein the value range of alpha is more than 0 and less than 1.
Step 3, obtaining a result S based on the adjacent dynamic property of the data periodic characteristics learned in the step 22iAnd the dynamic correction factor beta adjusts the prediction data S1i(ii) a Obtaining predicted data Si；
Here, Si＝S1i+β×S2i
Wherein the value range of beta is more than-0.5 and less than 0.5. Adjusting the predicted point S by a dynamic correction factor beta1iIncreasing accuracy of single value prediction and outputting predicted value Si。
Step 4, predicting data SiAssigned a value of wiFill in { Wi}w×1And delete { Wi}w×1The first data of (a) is to be processed,generating W of length Wi+1}w×1And re-executing step 1 to step 4 using the new input data.
Step 5, the sequence { S } with the length of p is obtained by realizing the circulation for the p timesp×1(ii) a Wherein, { S }p×1Each value in (3) is the prediction data S obtained in each cycle of the aforementioned step 3i。
Step 6 { S }p×1After passing through a Full Connected (FC) network, calculating a mixed loss function between the first network and the target sequence data, and training the first network, the second network and the FC network according to the calculation result of the mixed loss function so that the deviation degrees of the first network, the second network and the FC network and the target sequence data in values and time meet preset convergence conditions.
The above process is a network training process. When flow prediction is performed based on the first network, the second network, and the FC network shown in the figure, the sequence { S } may be obtained by performing the steps 1 to 5p×1Sequence { S }p×1Obtaining a flow prediction result (S) after FC network processingp×1。
Based on the foregoing embodiment, the embodiment of the present invention further provides a traffic prediction method. FIG. 3 is a flow chart illustrating a flow prediction method according to an embodiment of the present invention; as shown in fig. 3, the method includes:
step 201: obtaining first class data and second class data, wherein the first class data comprises a plurality of historical flow data before a predicted time point, the second class data comprises a plurality of historical flow data in the same period as the predicted time point, the plurality of second class data comprises the historical flow data of the time point in the same period as the predicted time point, and the historical flow data of the time point before and/or after the time point in the same period as the predicted time point;
step 202: obtaining initial prediction data based on the first type of data and a first network obtained by training, and obtaining adjustment data based on the second type of data and a second network obtained by training;
step 203: adjusting the initial prediction data according to the adjustment data to obtain prediction data;
step 204: and generating prediction sequence data based on the plurality of obtained prediction data, and processing the prediction sequence data through a full-connection network to obtain a flow prediction result at the prediction time point.
In this embodiment, the detailed processing procedure in step 201 to step 204 may refer to the description of the relevant steps in the foregoing network training method embodiment, and is not described herein again.
In this embodiment, the first network, the second network, and the fully-connected network are obtained by training based on the network training method described in the foregoing embodiment of the present application.
The embodiment of the invention also provides a network training device for flow prediction. Fig. 4 is a schematic structural diagram of a network training apparatus for traffic prediction according to an embodiment of the present invention, and as shown in fig. 4, the apparatus includes: a first processing unit 31, a second processing unit 32, an adjustment unit 33 and a training unit 34; wherein the content of the first and second substances,
the first processing unit 31 is configured to obtain a data set including a plurality of first data, and obtain first prediction data of a prediction time point based on the data set and a first network; the plurality of first data are historical flow data before the predicted time point;
the second processing unit 32 is configured to obtain a plurality of second data, and obtain adjustment data based on the plurality of second data and a second network, where the plurality of second data are historical traffic data in the same period as the predicted time point; the plurality of second data comprise historical flow data of time points which are in the same period as the predicted time point and also comprise historical flow data of time points before and/or after the time points which are in the same period as the predicted time point;
the adjusting unit 33 is configured to adjust the first prediction data according to the adjustment data and the dynamic adjustment parameter to obtain second prediction data, and generate prediction sequence data based on the obtained plurality of second prediction data;
the training unit 34 is configured to determine a loss with the target sequence data after the prediction sequence data is processed by the fully-connected network, and train the first network, the second network, and the fully-connected network according to the loss.
In some optional embodiments of the present invention, the first processing unit 31 is further configured to add the second prediction data to the data set, and delete one first data in the data set to obtain an updated data set; retrieving first predicted data for the predicted point in time based on the updated data set and the first network.
In some optional embodiments of the invention, the plurality of second data includes first historical flow data of a time point in a same period as the predicted time point, second historical flow data of a time point before the time point in the same period as the predicted time point, and third historical flow data of a time point after the time point in the same period as the predicted time point;
the second processing unit 32 is configured to obtain first adjustment data, second adjustment data, and third adjustment data corresponding to the first historical traffic data, the second historical traffic data, and the third historical traffic data, respectively, based on the second network; and performing weighting processing on the first adjustment data, the second adjustment data and the third adjustment data, and obtaining the adjustment data according to a weighting processing result.
In some optional embodiments of the present invention, the second processing unit 32 is configured to perform a weighted summation process on the first adjustment data, the second adjustment data, and the third adjustment data by using a dynamic adjustment parameter α as a weight of the first adjustment data, and using (1- α)/2 as weights of the second adjustment data and the third adjustment data, so as to obtain a weighted processing result.
In some optional embodiments of the present invention, the adjusting unit 33 is configured to obtain a correction value by a product of a dynamic correction factor β and the adjustment data, and add the correction value and the first prediction data to obtain the second prediction data.
In some optional embodiments of the present invention, the training unit 34 is configured to obtain third data in the predicted sequence data after being processed by the fully-connected network, and obtain fourth data in the target sequence data; the third data is the ith data in the predicted sequence data, and the fourth data is the ith data in the target sequence data; the value of i is greater than 0 and less than or equal to the number of data contained in the prediction sequence data or the target sequence data; determining a mean absolute percentage error between the third data and the fourth data, and determining a cosine distance between the third data and the fourth data, determining a loss based on the mean absolute percentage error and the cosine distance.
In the embodiment of the present invention, the first Processing Unit 31, the second Processing Unit 32, the adjusting Unit 33, and the training Unit 34 in the apparatus may be implemented by a Central Processing Unit (CPU), a Digital Signal Processor (DSP), a Micro Control Unit (MCU) or a Programmable Gate Array (FPGA) in practical application.
It should be noted that: in the network training device for traffic prediction according to the above embodiment, when performing network training, only the division of the program modules is illustrated, and in practical applications, the processing distribution may be completed by different program modules according to needs, that is, the internal structure of the device may be divided into different program modules to complete all or part of the processing described above. In addition, the network training device for traffic prediction and the network training method for traffic prediction provided in the above embodiments belong to the same concept, and specific implementation processes thereof are described in detail in the method embodiments and are not described herein again.
The embodiment of the invention also provides a flow prediction device. FIG. 5 is a schematic diagram of a flow prediction apparatus according to an embodiment of the present invention; as shown in fig. 5, the apparatus includes: a data acquisition unit 41 and a third processing unit 42; wherein the content of the first and second substances,
the data obtaining unit 41 is configured to obtain first class data and second class data, where the first class data includes a plurality of pieces of historical traffic data before a predicted time point, the second class data includes a plurality of pieces of historical traffic data in a same period as the predicted time point, and the plurality of pieces of second data includes the pieces of historical traffic data at a time point in the same period as the predicted time point and also includes pieces of historical traffic data at time points before and/or after the time point in the same period as the predicted time point;
the third processing unit 42 is configured to obtain initial prediction data based on the first type of data and a first network obtained through training; obtaining adjustment data based on the second type of data and a second network obtained through training; adjusting the initial prediction data according to the adjustment data and the dynamic adjustment parameters to obtain prediction data; and generating prediction sequence data based on the plurality of obtained prediction data, and processing the prediction sequence data through a full-connection network to obtain a flow prediction result at the prediction time point.
In the embodiment of the present invention, the data obtaining unit 41 and the third processing unit 42 in the apparatus can be implemented by a CPU, a DSP, an MCU, or an FPGA in practical application.
It should be noted that: in the flow prediction device provided in the above embodiment, when performing flow prediction, only the division of the program modules is illustrated, and in practical applications, the processing distribution may be completed by different program modules according to needs, that is, the internal structure of the device may be divided into different program modules to complete all or part of the processing described above. In addition, the flow prediction apparatus and the flow prediction method provided in the above embodiments belong to the same concept, and specific implementation processes thereof are described in the method embodiments and are not described herein again.
An embodiment of the present invention further provides an electronic device, where the electronic device may include the network training device or the traffic prediction device for traffic prediction in the foregoing embodiments. Fig. 6 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present invention, as shown in fig. 6, the electronic device includes a memory 52, a processor 51, and a computer program stored in the memory 52 and executable on the processor 51, and when the processor 51 executes the program, the processor 51 implements the steps of the network training method for traffic prediction according to the foregoing embodiment of the present invention, or when the processor 51 executes the program, the processor 51 implements the steps of the traffic prediction method according to the foregoing embodiment of the present invention.
Optionally, various components in the electronic device are coupled together by a bus system 53. It will be appreciated that the bus system 53 is used to enable communications among the components. The bus system 53 includes a power bus, a control bus, and a status signal bus in addition to the data bus. For clarity of illustration, however, the various buses are labeled as bus system 53 in fig. 6.
It will be appreciated that the memory 52 can be either volatile memory or nonvolatile memory, and can include both volatile and nonvolatile memory. Among them, the nonvolatile Memory may be a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a magnetic random access Memory (FRAM), a Flash Memory (Flash Memory), a magnetic surface Memory, an optical disk, or a Compact Disc Read-Only Memory (CD-ROM); the magnetic surface storage may be disk storage or tape storage. Volatile Memory can be Random Access Memory (RAM), which acts as external cache Memory. By way of illustration and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Enhanced Synchronous Dynamic Random Access Memory (Enhanced DRAM), Synchronous Dynamic Random Access Memory (SLDRAM), Direct Memory (DRmb Access), and Random Access Memory (DRAM). The memory 52 described in connection with the embodiments of the invention is intended to comprise, without being limited to, these and any other suitable types of memory.
The method disclosed in the above embodiments of the present invention may be applied to the processor 51, or implemented by the processor 51. The processor 51 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware or instructions in the form of software in the processor 51. The processor 51 described above may be a general purpose processor, a DSP, or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, or the like. The processor 51 may implement or perform the methods, steps, and logic blocks disclosed in embodiments of the present invention. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of the method disclosed by the embodiment of the invention can be directly implemented by a hardware decoding processor, or can be implemented by combining hardware and software modules in the decoding processor. The software modules may be located in a storage medium located in the memory 52, and the processor 51 reads the information in the memory 52 and performs the steps of the aforementioned method in conjunction with its hardware.
In an exemplary embodiment, the electronic Device may be implemented by one or more Application Specific Integrated Circuits (ASICs), DSPs, Programmable Logic Devices (PLDs), Complex Programmable Logic Devices (CPLDs), FPGAs, general purpose processors, controllers, MCUs, microprocessors (microprocessors), or other electronic components for performing the aforementioned methods.
In an exemplary embodiment, the present invention further provides a computer readable storage medium, such as a memory 52, comprising a computer program, which is executable by a processor 51 of an electronic device to perform the steps of the aforementioned method. The computer readable storage medium can be Memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface Memory, optical disk, or CD-ROM; or may be various devices including one or any combination of the above memories.
Embodiments of the present invention also provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the network training method for traffic prediction according to the foregoing embodiments of the present invention, or the computer program, when executed by the processor, implements the steps of the traffic prediction method according to the foregoing embodiments of the present invention.
The methods disclosed in the several method embodiments provided in the present application may be combined arbitrarily without conflict to obtain new method embodiments.
Features disclosed in several of the product embodiments provided in the present application may be combined in any combination to yield new product embodiments without conflict.
The features disclosed in the several method or apparatus embodiments provided in the present application may be combined arbitrarily, without conflict, to arrive at new method embodiments or apparatus embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only one logical function division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all the functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
Those of ordinary skill in the art will understand that: all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: a removable storage device, a ROM, a RAM, a magnetic or optical disk, or various other media that can store program code.
Alternatively, the integrated unit of the present invention may be stored in a computer-readable storage medium if it is implemented in the form of a software functional module and sold or used as a separate product. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or a part contributing to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, a ROM, a RAM, a magnetic or optical disk, or various other media that can store program code.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (18)
1. A method of network training for traffic prediction, the method comprising:
obtaining a data set containing a plurality of first data, and obtaining first prediction data of a prediction time point based on the data set and a first network; the plurality of first data are historical flow data before the predicted time point;
obtaining a plurality of second data, and obtaining adjustment data based on the plurality of second data and a second network, wherein the plurality of second data are historical flow data in the same period as the predicted time point; the plurality of second data comprise historical flow data of time points which are in the same period as the predicted time point and also comprise historical flow data of time points before and/or after the time points which are in the same period as the predicted time point;
adjusting the first prediction data according to the adjustment data and the dynamic adjustment parameters to obtain second prediction data, and generating prediction sequence data based on the obtained plurality of second prediction data;
and after the prediction sequence data is processed by a fully-connected network, determining loss with target sequence data, and training the first network, the second network and the fully-connected network according to the loss.
2. The method of claim 1, further comprising: adding the second prediction data to the data set, and deleting one first data in the data set to obtain an updated data set;
retrieving first predicted data for the predicted point in time based on the updated data set and the first network.
3. The method according to claim 1, wherein the plurality of second data includes first historical flow data of a time point in a same period as the predicted time point, second historical flow data of a time point previous to the time point in the same period as the predicted time point, and third historical flow data of a time point subsequent to the time point in the same period as the predicted time point;
obtaining adjustment data based on the plurality of second data and the second network includes:
respectively obtaining first adjustment data, second adjustment data and third adjustment data corresponding to the first historical traffic data, the second historical traffic data and the third historical traffic data based on the second network;
and performing weighting processing on the first adjustment data, the second adjustment data and the third adjustment data, and obtaining the adjustment data according to a weighting processing result.
4. The method of claim 3, wherein weighting the first adjustment data, the second adjustment data, and the third adjustment data comprises:
and taking a dynamic adjustment parameter alpha as a weight of the first adjustment data, and taking (1-alpha)/2 as weights of the second adjustment data and the third adjustment data, and performing weighted summation processing on the first adjustment data, the second adjustment data and the third adjustment data to obtain a weighted processing result.
5. The method of claim 1, wherein the adjusting the first prediction data according to the adjustment data and the dynamic adjustment parameter to obtain second prediction data comprises:
and obtaining a correction value by utilizing the product of the dynamic correction factor beta and the adjustment data, and adding the correction value and the first prediction data to obtain the second prediction data.
6. The method of claim 1, wherein determining the loss of the predicted sequence data from the target sequence data after processing the predicted sequence data through the fully-connected network comprises:
obtaining third data in the prediction sequence data processed by the full-connection network, and obtaining fourth data in the target sequence data; the third data is the ith data in the predicted sequence data, and the fourth data is the ith data in the target sequence data; the value of i is greater than 0 and less than or equal to the number of data contained in the prediction sequence data or the target sequence data;
determining a mean absolute percentage error between the third data and the fourth data, and determining a cosine distance between the third data and the fourth data, determining a loss based on the mean absolute percentage error and the cosine distance.
7. A method of traffic prediction, the method comprising:
obtaining first class data and second class data, wherein the first class data comprises a plurality of historical flow data before a predicted time point, the second class data comprises a plurality of historical flow data in the same period as the predicted time point, the plurality of second class data comprises the historical flow data of the time point in the same period as the predicted time point, and the historical flow data of the time point before and/or after the time point in the same period as the predicted time point;
obtaining initial prediction data based on the first type of data and a first network obtained by training, and obtaining adjustment data based on the second type of data and a second network obtained by training;
adjusting the initial prediction data according to the adjustment data and the dynamic adjustment parameters to obtain prediction data;
and generating prediction sequence data based on the plurality of obtained prediction data, and processing the prediction sequence data through a full-connection network to obtain a flow prediction result at the prediction time point.
8. The method of claim 7, wherein the first network, the second network, and the fully connected network are obtained based on the method training of any one of claims 1 to 6.
9. A network training apparatus for traffic prediction, the apparatus comprising: the device comprises a first processing unit, a second processing unit, an adjusting unit and a training unit; wherein the content of the first and second substances,
the first processing unit is used for obtaining a data set containing a plurality of first data and obtaining first prediction data of a prediction time point based on the data set and a first network; the plurality of first data are historical flow data before the predicted time point;
the second processing unit is configured to obtain a plurality of second data, and obtain adjustment data based on the plurality of second data and a second network, where the plurality of second data are historical traffic data in the same period as the predicted time point; the plurality of second data comprise historical flow data of time points which are in the same period as the predicted time point and also comprise historical flow data of time points before and/or after the time points which are in the same period as the predicted time point;
the adjusting unit is used for adjusting the first prediction data according to the adjustment data and the dynamic adjustment parameters to obtain second prediction data, and generating prediction sequence data based on the obtained second prediction data;
and the training unit is used for determining loss with target sequence data after the prediction sequence data is processed by a fully-connected network, and training the first network, the second network and the fully-connected network according to the loss.
10. The apparatus according to claim 9, wherein the first processing unit is further configured to add the second prediction data to the data set, and delete one first data in the data set to obtain an updated data set; retrieving first prediction data for the predicted point in time based on the updated data set and the first network.
11. The apparatus according to claim 10, wherein the plurality of second data includes first historical flow data of a time point in a same period as the predicted time point, second historical flow data of a time point previous to the time point in the same period as the predicted time point, and third historical flow data of a time point subsequent to the time point in the same period as the predicted time point;
the second processing unit is configured to obtain first adjustment data, second adjustment data, and third adjustment data corresponding to the first historical traffic data, the second historical traffic data, and the third historical traffic data, respectively, based on the second network; and performing weighting processing on the first adjustment data, the second adjustment data and the third adjustment data, and obtaining the adjustment data according to a weighting processing result.
12. The apparatus according to claim 11, wherein the second processing unit is configured to perform a weighted summation process on the first adjustment data, the second adjustment data, and the third adjustment data by using a dynamic adjustment parameter α as a weight of the first adjustment data, and using (1- α)/2 as weights of the second adjustment data and the third adjustment data, so as to obtain a weighted processing result.
13. The apparatus according to claim 9, wherein the adjusting unit is configured to obtain a correction value by multiplying a dynamic correction factor β by the adjustment data, and add the correction value and the first prediction data to obtain the second prediction data.
14. The apparatus of claim 9, wherein the training unit is configured to obtain third data in the predicted sequence data processed by the fully-connected network, and obtain fourth data in the target sequence data; the third data is the ith data in the predicted sequence data, and the fourth data is the ith data in the target sequence data; the value of i is greater than 0 and less than or equal to the number of data contained in the prediction sequence data or the target sequence data; determining a mean absolute percentage error between the third data and the fourth data, and determining a cosine distance between the third data and the fourth data, determining a loss based on the mean absolute percentage error and the cosine distance.
15. A flow prediction apparatus, characterized in that the apparatus comprises: a data acquisition unit and a third processing unit; wherein the content of the first and second substances,
the data acquisition unit is used for acquiring first-class data and second-class data, wherein the first-class data comprises a plurality of historical flow data before a predicted time point, the second-class data comprises a plurality of historical flow data in the same period as the predicted time point, the plurality of second data comprises the historical flow data at the same period as the predicted time point and the historical flow data at time points before and/or after the same period as the predicted time point;
the third processing unit is used for obtaining initial prediction data based on the first type of data and a first network obtained by training; obtaining adjustment data based on the second type of data and a second network obtained through training; adjusting the initial prediction data according to the adjustment data and the dynamic adjustment parameters to obtain prediction data; and generating prediction sequence data based on the plurality of obtained prediction data, and processing the prediction sequence data through a full-connection network to obtain a flow prediction result at the prediction time point.
16. The apparatus of claim 15, wherein the first network, the second network, and the fully connected network are obtained based on training of the network training apparatus for traffic prediction according to any one of claims 9 to 14.
17. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 6; alternatively, the program realizes the steps of the method of claim 7 or 8 when executed by a processor.
18. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the steps of the method of any one of claims 1 to 6 are implemented when the program is executed by the processor; alternatively, the processor implements the steps of the method of claim 7 or 8 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011284570.XA CN114584476A (en) | 2020-11-17 | 2020-11-17 | Traffic prediction method, network training device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011284570.XA CN114584476A (en) | 2020-11-17 | 2020-11-17 | Traffic prediction method, network training device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114584476A true CN114584476A (en) | 2022-06-03 |
Family
ID=81768735
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011284570.XA Pending CN114584476A (en) | 2020-11-17 | 2020-11-17 | Traffic prediction method, network training device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114584476A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115225546A (en) * | 2022-07-22 | 2022-10-21 | 北京天融信网络安全技术有限公司 | Method, device and equipment for predicting network flow |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107959640A (en) * | 2016-10-14 | 2018-04-24 | 腾讯科技（深圳）有限公司 | Network dispatching method and device |
| CN108924127A (en) * | 2018-06-29 | 2018-11-30 | 新华三信息安全技术有限公司 | A kind of generation method and device of flow baseline |
| CN109902886A (en) * | 2019-04-10 | 2019-06-18 | 广东电网有限责任公司 | A kind of two dimension power distribution communication net network flow prediction method |
| CN110858973A (en) * | 2018-08-23 | 2020-03-03 | ***通信集团山东有限公司 | Method and device for predicting network traffic of cell |
| US20200136975A1 (en) * | 2018-10-26 | 2020-04-30 | Hughes Network Systems, Llc | Monitoring a communication network |
-
2020
- 2020-11-17 CN CN202011284570.XA patent/CN114584476A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107959640A (en) * | 2016-10-14 | 2018-04-24 | 腾讯科技（深圳）有限公司 | Network dispatching method and device |
| CN108924127A (en) * | 2018-06-29 | 2018-11-30 | 新华三信息安全技术有限公司 | A kind of generation method and device of flow baseline |
| CN110858973A (en) * | 2018-08-23 | 2020-03-03 | ***通信集团山东有限公司 | Method and device for predicting network traffic of cell |
| US20200136975A1 (en) * | 2018-10-26 | 2020-04-30 | Hughes Network Systems, Llc | Monitoring a communication network |
| CN109902886A (en) * | 2019-04-10 | 2019-06-18 | 广东电网有限责任公司 | A kind of two dimension power distribution communication net network flow prediction method |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115225546A (en) * | 2022-07-22 | 2022-10-21 | 北京天融信网络安全技术有限公司 | Method, device and equipment for predicting network flow |
| CN115225546B (en) * | 2022-07-22 | 2023-11-28 | 北京天融信网络安全技术有限公司 | Network traffic prediction method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019238029A1 (en) | Convolutional neural network system, and method for quantifying convolutional neural network | |
| CN111638958B (en) | Cloud host load processing method and device, control equipment and storage medium | |
| CN110264270B (en) | Behavior prediction method, behavior prediction device, behavior prediction equipment and storage medium | |
| CN115085196B (en) | Power load predicted value determination method, device, equipment and computer readable medium | |
| CN110647974A (en) | Network layer operation method and device in deep neural network | |
| US20220138537A1 (en) | Probabilistic nonlinear relationships cross-multi time series and external factors for improved multivariate time series modeling and forecasting | |
| CN111783938A (en) | Time series prediction method and device | |
| CN116684330A (en) | Traffic prediction method, device, equipment and storage medium based on artificial intelligence | |
| CN115564152A (en) | Carbon emission prediction method and device based on STIRPAT model | |
| US10956811B2 (en) | Variable epoch spike train filtering | |
| CN114584476A (en) | Traffic prediction method, network training device and electronic equipment | |
| CN112085175A (en) | Data processing method and device based on neural network calculation | |
| CN114239949A (en) | Website access amount prediction method and system based on two-stage attention mechanism | |
| CN114064394A (en) | Safety monitoring method and device based on edge calculation and terminal equipment | |
| CN112561050B (en) | Neural network model training method and device | |
| CN112257958A (en) | Power saturation load prediction method and device | |
| JP2021033994A (en) | Text processing method, apparatus, device and computer readable storage medium | |
| US20210342691A1 (en) | System and method for neural time series preprocessing | |
| CN116490871A (en) | Automatically adjusting data access policies in data analysis | |
| CN112989203A (en) | Material delivery method, device, equipment and medium | |
| CN111626472A (en) | Scene trend judgment index computing system and method based on deep hybrid cloud model | |
| CN113435653B (en) | Method and system for predicting saturated power consumption based on logistic model | |
| CN114862003A (en) | Request quantity prediction method and device, electronic equipment and storage medium | |
| CN116684307A (en) | System capacity prediction method, device and equipment of Internet of things platform and storage medium | |
| CN117196050A (en) | Normal distributed quantum state preparation method and device, storage medium and electronic device |
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 |