Method, device, apparatus, and computer-readable medium for storing value-related information

Abstract

The embodiments of the present disclosure disclose a method, apparatus, device, and computer-readable medium for storing value-related information. A specific implementation of the method includes: obtaining a historical subject value processing data set and at least one value indicator; extracting at least one historical value indicator data sequence; classifying each historical indicator data at the same time; for each computing node, determining the remaining available computing resources and the node call order; determining the computing node information corresponding to each historical value indicator data group; for each historical time, executing a first generation step: determining the historical value indicator data group; calling at least one graphics processor to generate an indicator importance information group; calling a central processing unit to generate a current indicator importance information group; generating current value voucher information; and executing corresponding value voucher information storage. This implementation can efficiently and accurately generate value voucher information while fully utilizing online computing resources.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of computer technology, and in particular to a method, apparatus, device, and computer-readable medium for storing value-related information. Background Art

[0002] Currently, with the continuous development of society, the number of users and businesses is increasing. The value credentials (e.g., credit) of users and businesses are crucial. To determine the value credentials of users or businesses, the common approach is: first, obtain a historical value data sequence for the user or business. Then, using a pre-deployed image processor, this historical value data sequence is directly input into a value credential information generation model (e.g., a recurrent neural network model) to generate value credential information.

[0003] However, when using the above method to generate value certificate information, the following technical problems often occur:

[0004] When the historical value data series has a large data volume, using an image processor for online information generation often puts a lot of pressure on the processor to process data, forcing it to run under high load for a long time. When the temperature exceeds a predetermined threshold (e.g., 95 degrees), downtime is likely to occur. This leads to generation breakpoints in the value voucher information generation process, often resulting in the generated value voucher information being inefficient and inaccurate. In addition, using the historical value data series as the model input results in the generated value voucher information being inaccurate.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the Invention

[0006] The content of this disclosure is used to briefly introduce concepts that will be described in detail in the detailed description section below. The content of this disclosure is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0007] Some embodiments of the present disclosure propose value-related information storage methods, devices, equipment, and computer-readable media to solve one or more of the technical problems mentioned in the above background technology section.

[0008] In the first aspect, some embodiments of the present disclosure provide a method for storing value-related information, including: obtaining a historical subject value processing data set corresponding to a target subject and at least one value indicator related to value voucher information; extracting at least one historical value indicator data sequence corresponding to the at least one value indicator from the above historical subject value processing data set; classifying each historical indicator data in the at least one historical value indicator data sequence at the same time to generate a historical value indicator data group sequence, wherein each historical value indicator data in each historical value indicator data group corresponds to the same historical time; for each computing node in the current computing node cluster, determining the remaining available computing resources of at least one graphics processor deployed corresponding to the above computing node and the node calling order corresponding to the above computing node; according to the number of groups corresponding to the historical value indicator data group, the data byte size corresponding to the historical value indicator data group , the remaining available computing resources and node calling order corresponding to each computing node, determine the computing node information corresponding to each historical value indicator data group in the above-mentioned historical value indicator data group sequence; for each historical time in the historical time sequence, execute the first generation step: determine the historical value indicator data group corresponding to the above-mentioned historical time as the target historical value indicator data group; call at least one graphics processor in the computing node corresponding to the above-mentioned target historical value indicator data group to generate the corresponding indicator importance information group; according to the obtained indicator importance information group sequence, call the central processing unit to generate the current indicator importance information group for the above-mentioned target subject; according to the above-mentioned current indicator importance information group and the obtained current subject value processing data set, generate the current value voucher information; store the above-mentioned current value voucher information and the subject information corresponding to the above-mentioned target subject in the form of an encrypted key-value pair.

[0009] On the second aspect, some embodiments of the present disclosure provide a value-related information storage device, including: an acquisition unit, configured to acquire a historical subject value processing data set corresponding to a target subject and at least one value indicator related to value voucher information; an extraction unit, configured to extract at least one historical value indicator data sequence corresponding to the at least one value indicator from the above historical subject value processing data set; a classification unit, configured to classify each historical indicator data in the at least one historical value indicator data sequence at the same time to generate a historical value indicator data group sequence, wherein each historical value indicator data in each historical value indicator data group corresponds to the same historical time; a first determination unit, configured to determine, for each computing node in the current computing node cluster, the remaining available computing resources of at least one graphics processor deployed corresponding to the above computing node and the node call order corresponding to the above computing node; a second determination unit, configured to determine the number of groups corresponding to the historical value indicator data group, the number of historical value indicator data ... The data byte size, the remaining available computing resources corresponding to each computing node and the node calling order are used to determine the computing node information corresponding to each historical value indicator data group in the above-mentioned historical value indicator data group sequence; the execution unit is configured to execute the first generation step for each historical time in the historical time sequence: determine the historical value indicator data group corresponding to the above-mentioned historical time as the target historical value indicator data group; call at least one graphics processor in the computing node corresponding to the above-mentioned target historical value indicator data group to generate the corresponding indicator importance information group; the calling unit is configured to call the central processing unit according to the obtained indicator importance information group sequence to generate the current indicator importance information group for the above-mentioned target subject; the generation unit is configured to generate current value voucher information according to the above-mentioned current indicator importance information group and the obtained current subject value processing data set; the storage unit is configured to store the above-mentioned current value voucher information and the subject information corresponding to the above-mentioned target subject in the form of an encrypted key-value pair.

[0010] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0011] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner in the first aspect is implemented.

[0012] The above-described embodiments of the present disclosure have the following beneficial effects: Through the value-related information storage methods of some embodiments of the present disclosure, value voucher information can be efficiently and accurately generated while fully utilizing online computing resources. Specifically, the reason for the inaccuracy and efficiency of the relevant value voucher information is that, when the historical value data series corresponds to a large data volume, the use of an image processor for online information generation often places heavy data processing pressure on the processor, requiring prolonged operation under high load. When the temperature exceeds a predetermined threshold (e.g., 95 degrees), the processor is prone to crashes. This leads to generation breakpoints during the value voucher information generation process, often resulting in inefficient and inaccurate generated value voucher information. Furthermore, using the historical value data series as model input can result in inaccurate generated value voucher information. Based on this, the value-related information storage methods of some embodiments of the present disclosure first obtain a historical subject value processing dataset corresponding to the target subject and at least one value indicator related to the value voucher information to obtain a dataset corresponding to the value voucher information for primary analysis. Based on the at least one value indicator, the accuracy and speed of online value voucher information generation are improved while being more representative and reducing the data volume. Then, at least one historical value indicator data sequence corresponding to the at least one value indicator is extracted from the historical subject value processing data set to facilitate the subsequent generation of value voucher information. Next, each historical indicator data in the at least one historical value indicator data sequence is temporally classified to generate a sequence of historical value indicator data groups, wherein each historical value indicator data in each historical value indicator data group corresponds to the same historical time. Temporally classifying the historical indicator data facilitates the subsequent determination of the indicator importance at each historical time, thereby more accurately generating the indicator importance corresponding to each indicator from a temporal perspective. Next, for each compute node in the current compute node cluster, the remaining available computing resources of the at least one graphics processor deployed corresponding to the compute node and the node call order corresponding to the compute node are determined. By using a compute node cluster, with each compute node having multiple graphics processors deployed, the cluster allows for the rational allocation of available online resources without impacting other online tasks. By determining the remaining available computing resources and the node call order, the cluster prevents prolonged high load on the graphics processor, which could result in GPU downtime. Furthermore, based on the number of groups corresponding to the historical value indicator data group, the data byte size corresponding to the historical value indicator data group, the remaining available computing resources corresponding to each computing node and the node calling order, the computing node information corresponding to each historical value indicator data group in the above historical value indicator data group sequence is determined to achieve reasonable scheduling of resources, ensure full utilization of online resources, and greatly improve utilization efficiency.Secondly, for each historical time in the historical time series, the first generation step is performed: the first step is to determine the historical value indicator data group corresponding to the above historical time as the target historical value indicator data group, so as to facilitate the generation of indicator importance information. The second step is to call at least one graphics processor in the computing node corresponding to the above target historical value indicator data group to accurately generate the corresponding indicator importance information group. Furthermore, based on the obtained indicator importance information group sequence, the central processing unit is called to efficiently and accurately generate the current indicator importance information group for the above target subject. Thus, based on the above current indicator importance information group and the obtained current subject value processing data set, the current value voucher information can be generated under the condition of an accurate current indicator importance information group. Finally, the above current value voucher information and the subject information corresponding to the above target subject are stored in the form of an encrypted key-value pair to facilitate the corresponding use of the current value voucher information and the subject information. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other features, advantages, and aspects of the various embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0014] Figure 1 is a flowchart of some embodiments of the value-related information storage method according to the present disclosure;

[0015] Figure 2 is a schematic structural diagram of some embodiments of the value-related information storage device according to the present disclosure;

[0016] Figure 3 It is a structural diagram of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0017] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0018] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other.

[0019] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0020] It should be noted that the modifications of "one" and "multiple" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, they should be understood as "one or more".

[0021] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0022] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0023] refer to Figure 1 , shows a process 100 of some embodiments of the value-related information storage method according to the present disclosure. The value-related information storage method includes the following steps:

[0024] Step 101: Acquire a historical subject value processing data set corresponding to a target subject and at least one value indicator related to value voucher information.

[0025] In some embodiments, the subject (electronic device) executing the above-mentioned value-related information storage method can obtain a historical subject value processing dataset and at least one value indicator related to the value voucher information corresponding to a target subject via a wired or wireless connection. The target subject may be the subject for which value voucher information is to be generated. The value voucher information may be information representing the value voucher. In practice, in the rating field, the corresponding value voucher information may be the subject's creditworthiness. The historical subject value processing dataset may be a subject value processing dataset for a historical time period. The subject value processing dataset may be a dataset related to the target subject's corresponding value processing operations. In practice, value processing operations may include, but are not limited to, at least one of the following: subject debt repayment operations, subject borrowing operations, subject capital flow operations, and subject stock price change operations. The at least one value indicator may be a value indicator associated with the generation of the value voucher information. In practice, the at least one value indicator may be a pre-set indicator. In practice, the at least one value indicator may include, but is not limited to, at least one of the following: subject borrowing trend indicators, subject repayment trend indicators, subject financial indicators, subject input product consumption indicators, subject scale indicators, and subject operating indicators.

[0026] Step 102: extract at least one historical value indicator data sequence corresponding to the at least one value indicator from the historical subject value processing data set.

[0027] In some embodiments, the execution entity may extract at least one historical value indicator data sequence corresponding to the at least one value indicator from the historical entity value processing data set. There is a one-to-one correspondence between the value indicator in the at least one value indicator and the historical value indicator data sequence in the at least one historical value indicator data sequence. The historical value indicator data sequence may be the value indicator data of the target entity at corresponding historical times. That is, there is a one-to-one correspondence between the historical value indicator data in the historical value indicator data sequence and the historical time at each historical time.

[0028] As an example, the execution entity may utilize data extraction technology to extract at least one historical value indicator data sequence corresponding to the at least one value indicator from the historical entity value processing data set.

[0029] Step 103 , classifying the historical indicator data in the at least one historical value indicator data sequence according to time to generate a historical value indicator data group sequence.

[0030] In some embodiments, the execution entity may perform time classification on each historical indicator data in the at least one historical value indicator data sequence to generate a sequence of historical value indicator data groups, wherein each historical value indicator data in each historical value indicator data group corresponds to the same historical time.

[0031] Step 104 : For each computing node in the current computing node cluster, determine the remaining available computing resources of at least one graphics processor deployed corresponding to the computing node and the node call order corresponding to the computing node.

[0032] In some embodiments, for each compute node in the current compute node cluster, the execution entity may determine the remaining available computing resources of at least one graphics processing unit (GPU) deployed corresponding to the compute node and the node call order corresponding to the compute node. A compute node machine may include multiple compute nodes. A compute node may have at least one graphics processing unit (GPU) deployed corresponding to the compute node. The at least one GPU may support the execution of both online and offline tasks. That is, the at least one GPU may have both online and offline resources that can be scheduled. Online resources are used for online tasks, while offline resources are used for offline tasks. The remaining available computing resources may be the computing resources that the at least one GPU can support for online task execution. The node call order may be the order in which the compute nodes are used. The computing resources corresponding to each compute node may differ, and the compute node cluster also supports the addition of compute nodes. Computing nodes support the addition of GPUs to increase available computing resources. Computing nodes at the front of the node call order are called first, with higher support. Computing nodes may be servers used for resource calculation.

[0033] Here, by clustering computing nodes and setting the order of computing node calls, not only more computing resources are provided for online and offline tasks, but also the reasonable scheduling of computing resources can be dynamically achieved, avoiding long-term high loads on each graphics processor in a single computing node. At the same time, it also ensures the accurate and efficient generation of subsequent current indicator importance information groups.

[0034] As an example, the execution entity may determine the remaining available computing resources corresponding to the computing nodes and the node calling order by querying computing resources.

[0035] Step 105, determine the computing node information corresponding to each historical value indicator data group in the above historical value indicator data group sequence based on the number of groups corresponding to the historical value indicator data group, the data byte size corresponding to the historical value indicator data group, the remaining available computing resources corresponding to each computing node, and the node calling order.

[0036] In some embodiments, the execution entity may determine the computing node information corresponding to each historical value indicator data group in the historical value indicator data group sequence based on the number of groups corresponding to the historical value indicator data group, the data byte size corresponding to the historical value indicator data group, the remaining available computing resources corresponding to each computing node, and the node call order. The number of groups corresponding to the historical value indicator data group may be the amount of data in the historical value indicator data. The data byte size may be the storage byte size occupied by the historical value indicator data group. Specifically, the data byte size may be a value in bytes. The computing node corresponding to the historical value indicator data group may be the computing server to be processed for the historical value indicator data group.

[0037] In some optional implementations of some embodiments, determining the computing node information corresponding to each historical value indicator data group in the historical value indicator data group sequence based on the number of groups corresponding to the historical value indicator data group, the data byte size corresponding to the historical value indicator data group, the remaining available computing resources corresponding to each computing node, and the node calling order may include the following steps:

[0038] In the first step, the computing nodes are sorted according to the node call order to generate a computing node information sequence. The computing node information may be a node identifier of the computing node. The computing node information in the computing node information sequence may be arranged in the order of resource priority.

[0039] The second step is to determine the target percentile data byte of the data byte size sequence corresponding to the historical value indicator data group sequence. The data byte sizes in the data byte size sequence correspond one-to-one to the historical value indicator data groups in the historical value indicator data group sequence. The target percentile data byte may be the data byte whose size value of the corresponding data byte in the data byte size sequence falls below the target percentile. For example, the target percentile position may be 1 / 3.

[0040] The third step is to remove the computing node information corresponding to the remaining available computing resources that are less than the target percentile data bytes from the above computing node information sequence to obtain the computing node information sequence after removal.

[0041] The fourth step is to generate a sequence of remaining available computing resources for the above-removed computing node information sequence.

[0042] The fifth step is to determine the data byte size sequence corresponding to the historical value indicator data group sequence, wherein there is a one-to-one correspondence between the historical value indicator data groups in the historical value indicator data group sequence and the data byte sizes in the data byte size sequence.

[0043] In the sixth step, the above-mentioned data byte size sequence and the above-mentioned remaining available computing resource sequence are input into the computing node scheduling model to generate computing node scheduling information. The above-mentioned computing node scheduling information represents the correspondence between the data byte size subsequence and the remaining available computing resource subsequence, the sequence length corresponding to each data byte size subsequence is not uniform, and the total amount of data bytes corresponding to each data byte size sequence is less than the total amount of resources corresponding to the corresponding remaining available computing resource subsequence. The computing node scheduling model can be a model for node scheduling of computing nodes. The computing node scheduling information can represent information on how each computing node performs node scheduling. In practice, the computing node scheduling model can be a multi-layer fully connected layer connected in series.

[0044] The seventh step is to determine the computing node information corresponding to the above historical value indicator data group based on the above computing node scheduling information.

[0045] As an example, the execution entity may search for computing node information corresponding to the historical value indicator data group from computing node scheduling information.

[0046] Step 106: For each historical time in the historical time series, execute the first generation step:

[0047] Step 1061 : Determine the historical value indicator data group corresponding to the above historical time as the target historical value indicator data group.

[0048] In some embodiments, the execution entity may determine the historical value indicator data group corresponding to the historical time as the target historical value indicator data group.

[0049] As an example, the execution subject may determine the historical value indicator data group corresponding to the historical time as the target historical value indicator data group by means of data group query.

[0050] Step 1062: Call at least one graphics processor in the computing node corresponding to the target historical value indicator data group to generate a corresponding indicator importance information group.

[0051] In some embodiments, the execution entity may invoke at least one graphics processor in the computing node corresponding to the target historical value indicator data set to generate a corresponding indicator importance information set. The indicator importance information in the indicator importance information set corresponds one-to-one to the value indicator in the at least one value indicator. The indicator importance information may be a numerical value between 0 and 1. The higher the numerical value, the more important the indicator content representing the value indicator at the corresponding historical time is relative to the value voucher information.

[0052] In some optional implementations of some embodiments, calling at least one graphics processor in a computing node corresponding to the target historical value indicator data group to generate a corresponding indicator importance information group may include the following steps:

[0053] Call at least one graphics processor in the computing node corresponding to the target historical value indicator data group to perform the following second generation step:

[0054] Sub-step 1, obtaining at least one indicator overall evaluation information corresponding to the target subject at the above historical time, wherein the indicator overall evaluation information in the above at least one indicator overall evaluation information and the value indicator in the above at least one value indicator have a one-to-one correspondence. The indicator overall evaluation information can be the overall evaluation information of the indicator content at the historical time. The indicator overall evaluation information can be a numerical value between 0 and 100. The larger the numerical value, the more it represents the indicator tendency of the corresponding indicator content. For example, if the value indicator is the subject's borrowing trend indicator, the corresponding indicator overall evaluation information can be that the subject's borrowing amount is very large. For another example, if the value indicator is the subject's borrowing trend indicator, the corresponding indicator overall evaluation information can be the numerical value "80".

[0055] Sub-step 2: pre-processing each data in the target historical value indicator data group according to the overall evaluation information of the at least one indicator to obtain a pre-processed data group.

[0056] As an example, for each indicator overall evaluation information in at least one indicator overall evaluation information, first, the indicator data change information of the corresponding target historical value indicator data group is determined. Then, it is determined whether the indicator data change information matches the indicator overall evaluation information. Then, in response to determining that there is no match, data adjustments are performed on each data in the target historical value indicator data group based on the indicator overall evaluation information so that the indicator data change information corresponding to each indicator data after the adjustments matches the indicator overall evaluation information.

[0057] Sub-step 3: Obtain the first set of historical indicator importance information corresponding to the aforementioned historical time and the indicator importance information generation model for the current time. The indicator importance information generation model may be a neural network model used to generate indicator importance information for the current time. In practice, the indicator importance information generation model may be a multi-task classification model based on value credential information. For example, the indicator importance information generation model may comprise multiple concatenated convolutional layers, and the indicator importance information generation model may be trained using backpropagation.

[0058] Sub-step 4: Based on the preprocessed data set, using the indicator importance information generation model, generate a second historical indicator importance information set for the historical time period. The second historical indicator importance information in the second historical indicator importance information set has a one-to-one correspondence with a value indicator in the at least one value indicator.

[0059] As an example, the execution entity may input the pre-processed data group into the indicator importance information generation model to generate a second historical indicator importance information group.

[0060] Sub-step 5: generating the indicator importance information group according to the first historical indicator importance information group and the second historical indicator importance information group.

[0061] Optionally, generating the indicator importance information group according to the first historical indicator importance information group and the second historical indicator importance information group includes:

[0062] For each value indicator of the at least one value indicator, the following third generating step is performed:

[0063] Sub-step 1: determining the first historical indicator importance information and the second historical indicator importance information corresponding to the above value indicator, and using them as the first target historical indicator importance information and the second target historical indicator importance information respectively.

[0064] Sub-step 2: Determine a historical indicator importance information generation model corresponding to the above historical time. Each historical time has a corresponding historical indicator importance information generation model.

[0065] Sub-step 3: Obtain the model accuracy value corresponding to the above-mentioned historical indicator importance information generation model and the model accuracy value corresponding to the above-mentioned indicator importance information generation model, as the first model accuracy value and the above-mentioned second model accuracy value respectively.

[0066] Sub-step 4: normalize the first model accurate value and the second model accurate value to obtain a first normalized value and a second normalized value, wherein the sum of the first normalized value and the second normalized value is 1.

[0067] Sub-step 5: multiply the first normalized value and the first target historical indicator importance information to obtain first multiplication information.

[0068] Sub-step 6: multiply the second normalized value and the second target historical indicator importance information to obtain second multiplied information.

[0069] Sub-step 7: Add the first multiplication information and the second multiplication information to obtain the initial indicator importance information corresponding to the value indicator.

[0070] Sub-step 8: In response to determining that the initial indicator importance information is within the information interval corresponding to the value indicator, the initial indicator importance information is determined as the indicator importance information corresponding to the value indicator. Each value indicator has a corresponding indicator value interval (i.e., information interval). The indicator value interval can represent the upper and lower limits of the content corresponding to the value indicator.

[0071] Optionally, in response to determining that the initial indicator importance information is within the information interval corresponding to the value indicator, after determining the initial indicator importance information as the indicator importance information corresponding to the value indicator, the method further includes:

[0072] In the first step, in response to determining that the initial indicator importance information is greater than or equal to the maximum value corresponding to the information interval, the maximum value in the information interval is determined as the indicator importance information corresponding to the value indicator.

[0073] In the second step, in response to determining that the initial indicator importance information is less than or equal to the minimum value corresponding to the information interval, the minimum value in the information interval is determined as the indicator importance information corresponding to the value indicator.

[0074] Step 107: Based on the obtained indicator importance information group sequence, the central processing unit is called to generate a current indicator importance information group for the target subject.

[0075] In some embodiments, the execution entity may call a central processing unit (CPU) based on the obtained indicator importance information group sequence to generate a current indicator importance information group for the target entity, wherein the current indicator importance information may be an indicator importance information group at the current time.

[0076] In some optional implementations of some embodiments, calling a central processor based on the obtained indicator importance information group sequence to generate a current indicator importance information group for the target subject includes the following steps:

[0077] In the first step, each historical time in the historical time series is divided into time segments to generate a historical segment sequence, wherein a historical segment includes at least one adjacent historical time.

[0078] In the second step, at least one historical weight corresponding to each historical segment in the historical segment sequence is determined using at least one predetermined historical weight list, wherein the historical weight lists in the at least one historical weight list have a one-to-one correspondence with the value indicators in the at least one value indicator. Each historical weight list in the at least one historical weight list may be pre-set. The historical weight list may represent the correspondence between historical weights and historical segments under corresponding value indicators. The closer the historical segment is to the current time, the greater the corresponding historical weight.

[0079] In the third step, for each indicator importance information group in the indicator importance information group sequence, the following first information generation step is performed:

[0080] Sub-step 1: determine the historical segment corresponding to the above-mentioned indicator importance information group as the target historical segment.

[0081] Sub-step 2: determining at least one historical weight corresponding to the target historical frame as at least one target historical weight.

[0082] Sub-step 3: multiplying each indicator importance information in the indicator importance information group with each target history weight in at least one target history weight to generate a multiplied information group.

[0083] In the fourth step, for each value indicator in the at least one value indicator, the following second information generation step is performed:

[0084] Sub-step 1: Filter out the multiplication information sequence corresponding to the value indicator from the obtained multiplication information group sequence as the target multiplication information sequence.

[0085] Sub-step 2: Generate the first current indicator importance information through a linear regression model based on the above target multiplication information sequence.

[0086] The fifth step is to input the obtained multiplication information group sequence into the pre-trained current indicator importance information generation model to generate the second current indicator importance information group. Among them, the current indicator importance information generation model can be a neural network model that generates indicator importance information at the current time. In practice, each input of the current indicator importance information generation model is a multiplication information group, and the indicator importance information group at the current time (i.e., the second current indicator importance information group) is predicted by the multiplication information group at each historical time. Here, the current indicator importance information generation model can be a multi-layer convolution layer + fully connected layer connected in series. The current indicator importance information generation model aims to learn the overall indicator importance at each historical time, and through the overall changes in the overall indicator importance at each time, to learn the overall change characteristics, to predict the indicator importance of each value indicator at the current time (i.e., the second current indicator importance information group).

[0087] Step 6: Determine the first model weight corresponding to the linear regression model and the second model weight corresponding to the current indicator importance information generation model. The sum of the first model weight and the second model weight is 1. The second model weight is higher than the first model weight. The first model weight and the second model weight may be pre-set.

[0088] In the seventh step, based on the first model weight and the second model weight, the first current indicator importance information group and the second current indicator importance information group are weighted and summed to generate a weighted indicator importance information group.

[0089] The above-mentioned optional implementation methods and optional content in some embodiments, as an inventive point of the present disclosure, solve the technical problem of "the problem of insufficient accuracy of weighted indicator importance." Based on this, the present disclosure uses linear regression to predict the indicator importance of each value indicator at the current time. The model generated by the current indicator importance information can comprehensively predict the overall indicator importance at the current time. By combining the first model weight and the second model weight, the importance of at least one indicator corresponding to at least one value indicator at the current time can be accurately determined.

[0090] In some optional implementations of some embodiments, the above-mentioned calling a central processor according to the obtained indicator importance information group sequence to generate a current indicator importance information group for the above-mentioned target subject may include the following steps:

[0091] The first step is to call the above central processing unit to perform the following fourth generation step:

[0092] Sub-step 1: Generate an indicator importance information matrix corresponding to the indicator importance information group sequence. The corresponding row elements of the indicator importance information matrix represent historical time, and the corresponding column elements represent value indicator information.

[0093] Sub-step 2: performing element normalization on the indicator importance information matrix to obtain a normalized matrix, wherein the sum of the elements corresponding to each element in the normalized matrix is ​​1.

[0094] Sub-step 3: Generate at least one current indicator importance information corresponding to at least one value indicator.

[0095] As an example, generating at least one current indicator importance information corresponding to at least one value indicator includes:

[0096] For each value indicator of the at least one value indicator, the following fifth generating step is performed:

[0097] The first sub-step is to select from the normalized matrix a set of elements that are associated with the value indicators.

[0098] As an example, the execution entity may filter out the element set corresponding to the value indicator from the normalized matrix by searching for column elements.

[0099] The second sub-step is to add up the elements in the above element set to generate the current indicator importance information corresponding to the above value indicator.

[0100] Sub-step 4: determining the obtained at least one current indicator importance information as the above-mentioned current indicator importance information group.

[0101] Step 108: Generate current value voucher information based on the above-mentioned current indicator importance information group and the acquired current subject value processing data set.

[0102] In some embodiments, the execution subject may generate current value voucher information based on the current indicator importance information group and the acquired current subject value processing data set, wherein the current value voucher information may be value voucher information at the current time.

[0103] In some optional implementations of some embodiments, the above-mentioned process of generating the current value voucher information based on the above-mentioned current indicator importance information group and the acquired current subject value processing data set may include the following steps:

[0104] The first step is to extract at least one current value indicator data corresponding to the at least one value indicator from the current subject value processing data set, wherein the at least one current value indicator data may be indicator data corresponding to the at least one value indicator at the current time.

[0105] The second step is to perform weighted processing on the above-mentioned current indicator importance information group and the above-mentioned at least one current value indicator data to generate a weighted processing information group.

[0106] As an example, the execution entity may multiply each current indicator importance information with the corresponding current value indicator data to generate weighted processing information, thereby obtaining a weighted processing information group.

[0107] In the third step, the weighted information group is input into a pre-trained value voucher information generation model to generate the current value voucher information. The value voucher information generation model may be a neural network model that generates value voucher information. In practice, the value voucher information generation model may be a pre-trained time series classification model. The value voucher information generation model may be trained using backpropagation. The loss function corresponding to the value voucher information generation model may be a cross-entropy loss function. For example, the value voucher information generation model may be a long short-term memory (LSTM) model.

[0108] Step 109: Store the current value voucher information and the subject information corresponding to the target subject in the form of an encrypted key-value pair. The subject information may represent the subject identity information of the target subject. For example, the subject information may be the subject identifier of the target subject.

[0109] In some embodiments, the execution entity may store the current value voucher information and the entity information corresponding to the target entity in the form of an encrypted key-value pair.

[0110] As an example, first, the execution entity may use a target encryption algorithm to encrypt the current value credential information and the subject information to generate encrypted credential information and encrypted subject information. The encrypted credential information and encrypted subject information are then stored in the form of a key-value pair.

[0111] In some optional implementations of some embodiments, the steps further include:

[0112] The first step is to perform a credential rating on the current value credential information. The credential rating represents the target entity's credential level. In practice, the higher the credential rating, the higher the creditworthiness of the target entity.

[0113] The second step is to store the current value credential information, the credential rating, and the subject information in the form of encrypted key-value pairs. The specific implementation method will not be repeated here.

[0114] The above-described embodiments of the present disclosure have the following beneficial effects: Through the value-related information storage methods of some embodiments of the present disclosure, value voucher information can be efficiently and accurately generated while fully utilizing online computing resources. Specifically, the reason for the inaccuracy and efficiency of the relevant value voucher information is that, when the historical value data series corresponds to a large data volume, the use of an image processor for online information generation often places heavy data processing pressure on the processor, requiring prolonged operation under high load. When the temperature exceeds a predetermined threshold (e.g., 95 degrees), the processor is prone to crashes. This leads to generation breakpoints during the value voucher information generation process, often resulting in inefficient and inaccurate generated value voucher information. Furthermore, using the historical value data series as model input can result in inaccurate generated value voucher information. Based on this, the value-related information storage methods of some embodiments of the present disclosure first obtain a historical subject value processing dataset corresponding to the target subject and at least one value indicator related to the value voucher information to obtain a dataset corresponding to the value voucher information for primary analysis. Based on the at least one value indicator, the accuracy and speed of online value voucher information generation are improved while being more representative and reducing the data volume. Then, at least one historical value indicator data sequence corresponding to the at least one value indicator is extracted from the historical subject value processing data set to facilitate the subsequent generation of value voucher information. Next, each historical indicator data in the at least one historical value indicator data sequence is temporally classified to generate a sequence of historical value indicator data groups, wherein each historical value indicator data in each historical value indicator data group corresponds to the same historical time. Temporally classifying the historical indicator data facilitates the subsequent determination of the indicator importance at each historical time, thereby more accurately generating the indicator importance corresponding to each indicator from a temporal perspective. Next, for each compute node in the current compute node cluster, the remaining available computing resources of the at least one graphics processor deployed corresponding to the compute node and the node call order corresponding to the compute node are determined. By using a compute node cluster, with each compute node having multiple graphics processors deployed, the cluster allows for the rational allocation of available online resources without impacting other online tasks. By determining the remaining available computing resources and the node call order, the cluster prevents prolonged high load on the graphics processor, which could result in GPU downtime. Furthermore, based on the number of groups corresponding to the historical value indicator data group, the data byte size corresponding to the historical value indicator data group, the remaining available computing resources corresponding to each computing node and the node calling order, the computing node information corresponding to each historical value indicator data group in the above historical value indicator data group sequence is determined to achieve reasonable scheduling of resources, ensure full utilization of online resources, and greatly improve utilization efficiency.Secondly, for each historical time in the historical time series, the first generation step is performed: the first step is to determine the historical value indicator data group corresponding to the above historical time as the target historical value indicator data group, so as to facilitate the generation of indicator importance information. The second step is to call at least one graphics processor in the computing node corresponding to the above target historical value indicator data group to accurately generate the corresponding indicator importance information group. Furthermore, based on the obtained indicator importance information group sequence, the central processing unit is called to efficiently and accurately generate the current indicator importance information group for the above target subject. Thus, based on the above current indicator importance information group and the obtained current subject value processing data set, the current value voucher information can be generated under the condition of an accurate current indicator importance information group. Finally, the above current value voucher information and the subject information corresponding to the above target subject are stored in the form of an encrypted key-value pair to facilitate the corresponding use of the current value voucher information and the subject information.

[0115] Further references Figure 2 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a value-related information storage device. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the value-related information storage device can be specifically applied to various electronic devices.

[0116] like Figure 2As shown, a value-related information storage device 200 includes: an acquisition unit 201, an extraction unit 202, a classification unit 203, a first determination unit 204, a second determination unit 205, an execution unit 206, a calling unit 207, a generation unit 208 and a storage unit 209. Among them, the acquisition unit 201 is configured to acquire the historical subject value processing data set corresponding to the target subject and at least one value indicator related to the value voucher information; the extraction unit 202 is configured to extract at least one historical value indicator data sequence corresponding to the at least one value indicator from the above historical subject value processing data set; the classification unit 203 is configured to classify the various historical indicator data in the at least one historical value indicator data sequence at the same time to generate a historical value indicator data group sequence, wherein the historical time corresponding to each historical value indicator data in each historical value indicator data group is the same; the first determination unit 204 is configured to determine, for each computing node in the current computing node cluster, the remaining available computing resources of at least one graphics processor deployed corresponding to the above computing node and the node call order corresponding to the above computing node; the second determination unit 205 is configured to determine the number of groups corresponding to the historical value indicator data group, the data byte size corresponding to the historical value indicator data group, and each computing node. The corresponding remaining available computing resources and node calling order are used to determine the computing node information corresponding to each historical value indicator data group in the above-mentioned historical value indicator data group sequence; the execution unit 206 is configured to execute the first generation step for each historical time in the historical time sequence: determine the historical value indicator data group corresponding to the above-mentioned historical time as the target historical value indicator data group; call at least one graphics processor in the computing node corresponding to the above-mentioned target historical value indicator data group to generate the corresponding indicator importance information group; the calling unit 207 is configured to call the central processing unit according to the obtained indicator importance information group sequence to generate the current indicator importance information group for the above-mentioned target subject; the generating unit 208 is configured to generate the current value voucher information according to the above-mentioned current indicator importance information group and the obtained current subject value processing data set; the storage unit 209 is configured to store the above-mentioned current value voucher information and the subject information corresponding to the above-mentioned target subject in the form of an encrypted key-value pair.

[0117] It is understandable that the various units recorded in the value related information storage device 200 and the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the value-related information storage device 200 and the units contained therein, and will not be described in detail here.

[0118] Reference below Figure 3, which shows a structural schematic diagram of an electronic device (eg, an electronic device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0119] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. Various programs and data required for the operation of the electronic device 300 are also stored in the RAM 303. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0120] Typically, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as needed.

[0121] In particular, according to some embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from a network via the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are performed.

[0122] It should be noted that in some embodiments of the present disclosure, the computer-readable medium mentioned above may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or device. In some embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0123] In some embodiments, the client and server can communicate using any currently known or future developed network protocol, such as HTTP (HyperText Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0124] The above-mentioned computer-readable medium may be contained in the above-mentioned electronic device; or it may exist independently without being assembled into the electronic device. The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device: obtains the historical subject value processing data set corresponding to the target subject and at least one value indicator related to the value voucher information; extracts at least one historical value indicator data sequence corresponding to the above-mentioned at least one value indicator from the above-mentioned historical subject value processing data set; performs simultaneous classification on each historical indicator data in the above-mentioned at least one historical value indicator data sequence to generate a historical value indicator data group sequence, wherein each historical value indicator data in each historical value indicator data group corresponds to the same historical time; for each computing node in the current computing node cluster, determines the remaining available computing resources of at least one graphics processor deployed corresponding to the above-mentioned computing node and the node calling order corresponding to the above-mentioned computing node; according to the number of groups corresponding to the historical value indicator data group, the historical value indicator data group According to the corresponding data byte size, the remaining available computing resources corresponding to each computing node and the node calling order, the computing node information corresponding to each historical value indicator data group in the above-mentioned historical value indicator data group sequence is determined; for each historical time in the historical time sequence, the first generation step is executed: the historical value indicator data group corresponding to the above-mentioned historical time is determined as the target historical value indicator data group; at least one graphics processor in the computing node corresponding to the above-mentioned target historical value indicator data group is called to generate the corresponding indicator importance information group; according to the obtained indicator importance information group sequence, the central processing unit is called to generate the current indicator importance information group for the above-mentioned target subject; according to the above-mentioned current indicator importance information group and the obtained current subject value processing data set, the current value voucher information is generated; the above-mentioned current value voucher information and the subject information corresponding to the above-mentioned target subject are stored in the form of an encrypted key-value pair.

[0125] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0126] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0127] The units described in some embodiments of the present disclosure may be implemented in software or in hardware. The units described may also be provided in a processor. For example, they may be described as follows: a processor including an acquisition unit, an extraction unit, a classification unit, a first determination unit, a second determination unit, an execution unit, a call unit, a generation unit, and a storage unit. The names of these units do not, in some cases, constitute limitations on the units themselves. For example, the acquisition unit may also be described as a "unit for acquiring a historical subject value processing data set corresponding to a target subject and at least one value indicator related to value credential information."

[0128] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0129] The above description is only an illustration of some preferred embodiments of the present disclosure and the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A method for storing value-related information, comprising: Obtaining a historical subject value processing data set corresponding to the target subject and at least one value indicator related to the value voucher information; extracting at least one historical value indicator data sequence corresponding to the at least one value indicator from the historical subject value processing data set; Classifying each historical indicator data in the at least one historical value indicator data sequence by time to generate a historical value indicator data group sequence, wherein each historical value indicator data in each historical value indicator data group corresponds to the same historical time; For each computing node in the current computing node cluster, determining the remaining available computing resources of at least one graphics processor deployed corresponding to the computing node and the node call order corresponding to the computing node; Determine the computing node information corresponding to each historical value indicator data group in the historical value indicator data group sequence according to the number of groups corresponding to the historical value indicator data group, the data byte size corresponding to the historical value indicator data group, the remaining available computing resources corresponding to each computing node, and the node calling order, including: sorting the various computing nodes according to the node calling order to generate a computing node information sequence; determining the target percentile data byte of the data byte size sequence corresponding to the historical value indicator data group sequence; removing the computing node information whose corresponding remaining available computing resources are less than the target percentile data byte from the computing node information sequence to obtain the computing node information sequence after removal; generating the remaining available computing resource sequence for the computing node information sequence after removal; determining the data byte size sequence corresponding to the historical value indicator data group sequence; inputting the data byte size sequence and the remaining available computing resource sequence into a computing node scheduling model to generate computing node scheduling information; and determining the computing node information corresponding to the historical value indicator data group according to the computing node scheduling information; For each historical time in the historical time series, perform the first generation step: Determining a historical value indicator data group corresponding to the historical time as a target historical value indicator data group; Invoking at least one graphics processor in a computing node corresponding to the target historical value indicator data group to generate a corresponding indicator importance information group; According to the obtained indicator importance information group sequence, a central processing unit is called to generate a current indicator importance information group for the target subject; Generate current value voucher information based on the current indicator importance information group and the acquired current subject value processing data set; Storing the current value credential information and the subject information corresponding to the target subject in the form of an encrypted key-value pair; The method of calling a central processor according to the obtained indicator importance information group sequence to generate a current indicator importance information group for the target subject includes: Performing time division on each historical time in the historical time series to generate a historical fragment sequence; Determining at least one historical weight corresponding to each historical segment in the historical segment sequence by using at least one predetermined historical weight list; For each indicator importance information group in the indicator importance information group sequence, the following first information generation step is performed: determining a historical segment corresponding to the indicator importance information group as a target historical segment; determining at least one historical weight corresponding to the target historical segment as at least one target historical weight; multiplying each indicator importance information in the indicator importance information group by each target historical weight in the at least one target historical weight to generate a multiplied information group; For each value indicator in the at least one value indicator, the following second information generation step is performed: a multiplication information sequence corresponding to the value indicator is selected from the obtained multiplication information group sequence as a target multiplication information sequence; and first current indicator importance information is generated based on the target multiplication information sequence using a linear regression model; Inputting the obtained multiplied information group sequence into a pre-trained current indicator importance information generation model to generate a second current indicator importance information group; Determine a first model weight corresponding to the linear regression model and a second model weight corresponding to the current indicator importance information generation model; According to the first model weight and the second model weight, a weighted sum is performed on the first current indicator importance information group and the second current indicator importance information group to generate a weighted indicator importance information group.

2. The method according to claim 1, wherein The calling of at least one graphics processor in a computing node corresponding to the target historical value indicator data group to generate a corresponding indicator importance information group includes: Call at least one graphics processor in a computing node corresponding to the target historical value indicator data group to perform the following second generation step: Obtaining at least one indicator overall evaluation information corresponding to the target subject at the historical time, wherein the indicator overall evaluation information in the at least one indicator overall evaluation information and the value indicator in the at least one value indicator have a one-to-one correspondence; performing data preprocessing on each data in the target historical value indicator data group according to the at least one overall indicator evaluation information to obtain a preprocessed data group; Obtaining a first historical indicator importance information group corresponding to the historical time and an indicator importance information generation model at the current time; Based on the preprocessed data group, using the indicator importance information generation model to generate a second historical indicator importance information group at the historical time; The indicator importance information group is generated according to the first historical indicator importance information group and the second historical indicator importance information group.

3. The method according to claim 2, wherein: Generating the indicator importance information group according to the first historical indicator importance information group and the second historical indicator importance information group includes: For each value indicator of the at least one value indicator, the following third generating step is performed: Determine the first historical indicator importance information and the second historical indicator importance information corresponding to the value indicator, as the first target historical indicator importance information and the second target historical indicator importance information respectively; Determine a historical indicator importance information generation model corresponding to the historical time; Obtaining a model accuracy value corresponding to the historical indicator importance information generation model and a model accuracy value corresponding to the indicator importance information generation model, as a first model accuracy value and a second model accuracy value, respectively; Normalizing the first model accurate value and the second model accurate value to obtain a first normalized value and a second normalized value; Multiplying the first normalized value and the first target historical indicator importance information to obtain first multiplication information; Multiplying the second normalized value and the second target historical indicator importance information to obtain second multiplied information; Adding the first multiplication information and the second multiplication information to obtain initial indicator importance information corresponding to the value indicator; In response to determining that the initial indicator importance information is within the information interval corresponding to the value indicator, the initial indicator importance information is determined as the indicator importance information corresponding to the value indicator.

4. The method according to claim 3, wherein: After, in response to determining that the initial indicator importance information is within the information interval corresponding to the value indicator, determining the initial indicator importance information as the indicator importance information corresponding to the value indicator, the method further includes: In response to determining that the initial indicator importance information is greater than or equal to the maximum value corresponding to the information interval, determining the maximum value in the information interval as the indicator importance information corresponding to the value indicator; In response to determining that the initial indicator importance information is less than or equal to the minimum value corresponding to the information interval, the minimum value in the information interval is determined as the indicator importance information corresponding to the value indicator.

5. The method according to claim 1, wherein The computing node scheduling information represents the correspondence between the data byte size subsequences and the remaining available computing resource subsequences. The sequence lengths corresponding to the various data byte size subsequences are not uniform, and the total amount of data bytes corresponding to each data byte size sequence is less than the total amount of resources corresponding to the corresponding remaining available computing resource subsequence.

6. The method according to claim 5, wherein: The method of calling a central processor according to the obtained indicator importance information group sequence to generate a current indicator importance information group for the target subject includes: The central processing unit is called to execute the following fourth generation step: Generate an indicator importance information matrix corresponding to the indicator importance information group sequence; Performing element normalization on the indicator importance information matrix to obtain a normalized matrix; Generate at least one current indicator importance information corresponding to at least one value indicator; The obtained at least one current indicator importance information is determined as the current indicator importance information group.

7. The method according to claim 1, wherein The step of processing the data set based on the current indicator importance information group and the acquired current subject value to generate current value voucher information includes: extracting at least one current value indicator data corresponding to the at least one value indicator from the current subject value processing data set; performing weighting processing on the current indicator importance information group and the at least one current value indicator data to generate a weighted processing information group; The weighted processed information group is input into a pre-trained value voucher information generation model to generate the current value voucher information.

8. A value-related information storage device comprising: An acquisition unit configured to acquire a historical subject value processing data set corresponding to a target subject and at least one value indicator related to the value voucher information; An extraction unit configured to extract at least one historical value indicator data sequence corresponding to the at least one value indicator from the historical subject value processing data set; a classification unit configured to perform simultaneous classification on each historical indicator data in the at least one historical value indicator data sequence to generate a sequence of historical value indicator data groups, wherein each historical value indicator data in each historical value indicator data group corresponds to the same historical time; A first determining unit is configured to determine, for each computing node in the current computing node cluster, remaining available computing resources of at least one graphics processor deployed corresponding to the computing node and a node call order corresponding to the computing node; The second determining unit is configured to determine the computing node information corresponding to each historical value indicator data group in the historical value indicator data group sequence according to the number of groups corresponding to the historical value indicator data group, the data byte size corresponding to the historical value indicator data group, the remaining available computing resources corresponding to each computing node, and the node calling order, including: sorting each computing node according to the node calling order to generate a computing node information sequence; determining the target percentile data byte of the data byte size sequence corresponding to the historical value indicator data group sequence; removing the computing node information whose corresponding remaining available computing resources are less than the target percentile data byte from the computing node information sequence to obtain the computing node information sequence after removal; generating the remaining available computing resource sequence for the computing node information sequence after removal; determining the data byte size sequence corresponding to the historical value indicator data group sequence; inputting the data byte size sequence and the remaining available computing resource sequence into a computing node scheduling model to generate computing node scheduling information; and determining the computing node information corresponding to the historical value indicator data group according to the computing node scheduling information; The execution unit is configured to, for each historical time in the historical time series, perform a first generation step: determine a historical value indicator data group corresponding to the historical time as a target historical value indicator data group; call at least one graphics processor in a computing node corresponding to the target historical value indicator data group to generate a corresponding indicator importance information group; a calling unit configured to call a central processor according to the obtained indicator importance information group sequence to generate a current indicator importance information group for the target subject; a generating unit configured to generate current value voucher information based on the current indicator importance information group and the acquired current subject value processing data set; a storage unit configured to store the current value credential information and the subject information corresponding to the target subject in the form of an encrypted key-value pair; The method of calling a central processor according to the obtained indicator importance information group sequence to generate a current indicator importance information group for the target subject includes: Performing time division on each historical time in the historical time series to generate a historical fragment sequence; Determining at least one historical weight corresponding to each historical segment in the historical segment sequence by using at least one predetermined historical weight list; For each indicator importance information group in the indicator importance information group sequence, the following first information generation step is performed: determining a historical segment corresponding to the indicator importance information group as a target historical segment; determining at least one historical weight corresponding to the target historical segment as at least one target historical weight; multiplying each indicator importance information in the indicator importance information group by each target historical weight in the at least one target historical weight to generate a multiplied information group; For each value indicator in the at least one value indicator, the following second information generation step is performed: a multiplication information sequence corresponding to the value indicator is selected from the obtained multiplication information group sequence as a target multiplication information sequence; and first current indicator importance information is generated based on the target multiplication information sequence using a linear regression model; Inputting the obtained multiplied information group sequence into a pre-trained current indicator importance information generation model to generate a second current indicator importance information group; Determine a first model weight corresponding to the linear regression model and a second model weight corresponding to the current indicator importance information generation model; According to the first model weight and the second model weight, a weighted sum is performed on the first current indicator importance information group and the second current indicator importance information group to generate a weighted indicator importance information group.

9. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer-readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images