CN110634036A

CN110634036A - Method and device for distributing configuration objects, storage medium and electronic equipment

Info

Publication number: CN110634036A
Application number: CN201910927744.0A
Authority: CN
Inventors: 张佳伟
Original assignee: Golden Melon Seed Technology Development (beijing) Co Ltd
Current assignee: Golden Melon Seed Technology Development (beijing) Co Ltd
Priority date: 2019-09-27
Filing date: 2019-09-27
Publication date: 2019-12-31

Abstract

The invention provides a method, a device, a storage medium and an electronic device for distributing configuration objects, wherein the method comprises the following steps: presetting an experiment rule; acquiring a target object and determining object parameters of the target object; determining an effective experiment rule corresponding to the target object in the experiment rules; determining first actual quantity ratios distributed to different models, determining a model with the minimum current ratio according to the first actual quantity ratios and first standard quantity ratios of effective experiment rules, and distributing target objects to the model with the minimum current ratio. By the method, the device, the storage medium and the electronic equipment for distributing the configuration objects, the configuration objects can be distributed in real time based on the experiment rule, the real-time performance is good, and the experiment range is large; the object distribution is configured to the model with the least current occupation ratio, the object can be uniformly distributed to different models, uniform distribution is realized, and the local and overall states can gradually reach the optimal state.

Description

Method and device for distributing configuration objects, storage medium and electronic equipment

Technical Field

The present invention relates to the field of object processing technologies, and in particular, to a method and an apparatus for allocating objects in a streaming manner, a storage medium, and an electronic device.

Background

At present, when the used cars need to be priced or adjusted, the prices of the used cars can be determined manually, and the used cars can also be priced or adjusted based on car selling models. In order to optimize pricing or price adjustment results of the car selling model, the traditional scheme generally selects part of vehicles from part of cities at regular time, uses half of the vehicles as model vehicles, determines prices of the model vehicles based on the car selling model, uses the other half as comparison vehicles, and determines prices of the comparison vehicles in a manual intervention mode; and optimizing the car selling model by comparing the price difference between the model car and the comparison car.

The traditional scheme can only select part of vehicles, and the experimental range is small; the real-time performance is poor by adopting a mode of selecting vehicles at regular time and comparing; by adopting a simple 'folding and half-splitting' method, the situation that the data cannot be equally split can be met, and the sample data is simple and cannot be adapted to further requirements.

Disclosure of Invention

To solve the above problems, embodiments of the present invention provide a method, an apparatus, a storage medium, and an electronic device for distributing configuration objects.

In a first aspect, an embodiment of the present invention provides a method for distributing configuration objects, including:

presetting an experiment rule, wherein the experiment rule comprises experiment object parameters and a first standard quantity ratio for distributing objects to different models;

acquiring a target object and determining object parameters of the target object;

determining an effective experiment rule corresponding to the target object in the experiment rules according to the object parameters of the target object, wherein the effective experiment rule is an experiment rule with experiment object parameters matched with the object parameters of the target object;

determining a first actual quantity ratio of objects corresponding to the effective experiment rules to be distributed to different models, determining a model with the minimum current ratio according to the first actual quantity ratio and a first standard quantity ratio of the effective experiment rules, and distributing the target objects to the model with the minimum current ratio.

In a second aspect, an embodiment of the present invention further provides an apparatus for distributing configuration objects, including:

the rule setting module is used for presetting an experiment rule, wherein the experiment rule comprises experiment object parameters and a first standard quantity ratio for distributing objects to different models;

the acquisition module is used for acquiring a target object and determining object parameters of the target object;

the rule matching module is used for determining an effective experiment rule corresponding to the target object in the experiment rules according to the object parameters of the target object, wherein the effective experiment rule is an experiment rule with experiment object parameters matched with the object parameters of the target object;

and the distribution model module is used for determining a first actual quantity ratio of objects corresponding to the effective experiment rule to be distributed to different models, determining a model with the minimum current ratio according to the first actual quantity ratio and a first standard quantity ratio of the effective experiment rule, and distributing the target objects to the model with the minimum current ratio.

In a third aspect, an embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores computer-executable instructions, where the computer-executable instructions are used in any one of the foregoing methods for offloading a configuration object.

In a fourth aspect, an embodiment of the present invention further provides an electronic device, including:

at least one processor; and the number of the first and second groups,

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any of the methods of shunting configuration objects described above.

In the solution provided by the first aspect of the embodiment of the present invention, an experiment rule is preset, and when a target object to be configured by shunting is obtained, the target object is allocated based on the experiment rule, so that the target object can be configured by shunting in real time, and the method has good real-time performance and a large experiment range; meanwhile, the target object is distributed to the model with the least current occupation ratio, the object can be uniformly distributed to different models, uniform distribution is realized, and the local part and the whole part can gradually reach the optimal state. In addition, the objects are distributed to the corresponding models to be processed, the processing results are convenient to count, the processing results of different models can be compared, and the models can be conveniently trained and optimized subsequently.

In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a flowchart illustrating a method for distributing configuration objects according to an embodiment of the present invention;

fig. 2 is a flowchart illustrating a specific method for determining an effective experimental rule in the method for offloading the configuration object according to the embodiment of the present invention;

fig. 3 is a schematic structural diagram illustrating an apparatus for distributing configuration objects according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of an electronic device for performing a method of shunting a configuration object according to an embodiment of the present invention.

Detailed Description

In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", and the like, indicate orientations and positional relationships based on those shown in the drawings, and are used only for convenience of description and simplicity of description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be considered as limiting the present invention.

Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.

In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," "secured," and the like are to be construed broadly and can, for example, be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.

An embodiment of the present invention provides a method for allocating objects in a streaming manner, which is shown in fig. 1, and includes:

step 101: and presetting an experiment rule, wherein the experiment rule comprises experiment object parameters and a first standard quantity ratio for distributing the object to different models.

In the embodiment of the invention, an experimental rule for distributing objects is preset and maintained, wherein the experimental rule comprises content related to the objects, namely experimental object parameters; meanwhile, the scheme provided by this embodiment is further provided with a plurality of models capable of processing the objects, and the experimental rule further includes content for allocating the objects, that is, a first standard quantity proportion, which is used to preset the quantity proportion of the objects to be allocated by different models. The object in this embodiment refers to a thing that needs to be assigned to the corresponding model for processing, such as a car source that needs to be priced. In this embodiment, different models refer to multiple models that can process an object, and may be different versions of the same model. Preferably, different models in this embodiment refer to models with different version numbers, and the models can be optimized better subsequently based on processing results of the different models; correspondingly, the first standard quantity ratio at this time is the ratio of the quantity of the objects allocated to the models of different versions.

Specifically, the subject parameter corresponds to a multidimensional parameter of the subject, i.e., the subject parameter can describe the subject defined to conform to the subject parameter from a multidimensional perspective. For example, if the subject is a vehicle source to be priced, the subject parameters may include: source city, price interval, train, etc. And determining which experimental rule or rules the object corresponds to by judging whether the attribute parameters of the object are matched with the parameters of the experimental object. In addition, the subject parameters may also include more parameters, such as car source color, car age, etc.; and for different types of subjects, the subject parameters can be changed accordingly. The experiment rule in this embodiment is used for shunting configuration objects, and even if any object is acquired, the allocation can be performed, the experiment range of the object is large, and the allocation can be performed in time.

Step 102: and acquiring a target object and determining object parameters of the target object.

In the embodiment of the invention, the target objects are the objects needing to be distributed, and each object has corresponding attribute parameters, namely object parameters. Since the object parameters of the target object may include multiple items, in this embodiment, only the object parameters related to the experiment object parameters in the experiment rule may be determined. For example, the experimental object parameters include a vehicle source city, a price interval, and a vehicle series, and accordingly, only the city, the price, and the vehicle series of the target object may be determined in step 102.

Step 103: an effective experiment rule corresponding to the target object is determined in the experiment rules according to the object parameters of the target object, wherein the effective experiment rule is an experiment rule with experiment object parameters matched with the object parameters of the target object.

In the embodiment of the invention, whether the experimental rule corresponds to the target object can be determined according to whether the experimental object parameters of the experimental rule are matched with the object parameters of the target object, namely whether the experimental rule is an effective experimental rule can be determined. In this embodiment, the experiment rule that all items of the parameters of the experiment object are matched with the object parameters of the corresponding items of the target object may be used as the effective experiment rule corresponding to the target object, or the experiment rule that the parameters of the experiment object are partially matched with the object parameters of the corresponding items of the target object may also be used as the effective experiment rule corresponding to the target object; the former are preferably all matched. Still taking the object as the car source to be priced as an example, the experimental object parameters of the experimental rule include a car source city, a price interval and a car system, and the experimental object parameters of the experimental rule are [ Beijing, Tianjin; 10 to 15 ten thousand; domestic vehicle ], the vehicle source city is Beijing or Tianjin, the vehicle source price is between 10 and 15 thousands, and the vehicle sources of the domestic vehicles all accord with the experimental object parameters, namely the experimental rules can be used as the effective experimental rules of the corresponding vehicle sources.

Optionally, a plurality of experiment rules may be preset, and since the experiment object parameters of different experiment rules may have an intersection, the same object may correspond to the plurality of experiment rules. And when the object parameters of the target object match the plurality of experiment rules, selecting one of the plurality of experiment rules as an effective experiment rule corresponding to the target object. Specifically, referring to fig. 2, the step 103 "determining an effective experiment rule corresponding to the target object in the experiment rule according to the object parameter of the target object" includes:

step 1031: all experimental rules having experimental subject parameters matching the subject parameters of the target subject are determined, the determined experimental rules being intermediate experimental rules.

In this embodiment, after a target object to be allocated is acquired, it is sequentially determined whether object parameters of the target object are matched with experimental object parameters of an experimental rule, and if the object parameters of the target object are matched with the experimental object parameters of the experimental rule, the experimental rule is an experimental rule corresponding to the target object, that is, an intermediate experimental rule; otherwise, no processing may be performed. And then, determining whether the object parameters of the target object are matched with the experimental object parameters of other experimental rules again until all the experimental rules are traversed, and determining all the intermediate experimental rules of the target object. Those skilled in the art will appreciate that the "intermediate experimental rules" and the "valid experimental rules" are both experimental rules in nature, and that the "intermediate experimental rules" and the "valid experimental rules" may be different names of the same experimental rule at different nodes.

Step 1032: and determining the matching degree of each intermediate experiment rule according to the experiment object parameters of the intermediate experiment rules, and taking the intermediate experiment rule with the highest matching degree as an effective experiment rule corresponding to the target object. The experimental object parameters comprise one or more of default item quantity, creation time and parameter range, and the lower the default item quantity of the intermediate experimental rule, the closer the creation time to the current time or the smaller the parameter range, the higher the matching degree of the intermediate experimental rule.

In the embodiment of the invention, the intermediate experiment rule with the highest matching degree is used as the experiment rule which is most matched with the object parameter of the target object, namely the effective experiment rule. Specifically, the parameters of the experimental subject include multiple parameters, each parameter may be set as a default parameter, and the "default number" represents the number set as the default in the parameters of the experimental subject; the "creation time" of the subject parameter is the time for creating the experimental rule; furthermore, each of the subject parameters may include multiple elements (e.g., the car source city may include beijing and tianjin), or include an interval (e.g., the car source price may be in the interval of 10 to 15 ten thousand), and the "parameter range" represents the range covered by the subject parameter.

In this embodiment, the smaller the number of default items of the intermediate experimental rule, the closer the creation time is to the current time, or the smaller the parameter range is, the higher the matching degree of the corresponding intermediate experimental rule is. Specifically, the smaller the number of default items, the more precise the range defined by the experimental rule is, the higher the matching degree with the corresponding object parameter is. For example, the vehicle source city and the vehicle series in the intermediate experiment rule a are both default items, that is, the vehicle source of any city and any vehicle series is matched with the intermediate experiment rule a, and the number of the default items of the intermediate experiment rule a is 2; only the vehicle source city in the intermediate experiment rule B is a default item, namely the number of the default items of the intermediate experiment rule B is 1; if a certain object is matched with both the intermediate experiment rules A and B, but the number of default items of the intermediate experiment rule B is less, the object is matched with the intermediate experiment rule B more, and the matching degree of the intermediate experiment rule B with the object is higher. Similarly, the closer the creation time of the intermediate experiment rule is to the current time, the newer the intermediate experiment rule is, the higher the real-time property is, and the corresponding target object should be assigned to the newer intermediate experiment rule, that is, the higher the matching degree of the intermediate experiment rule at this time is.

In addition, the smaller the parameter range is, the more accurate the intermediate experiment rule hits the target object, and the higher the matching degree between the intermediate experiment rule and the target object is. For example, if the car source city in the intermediate experimental rule a is beijing and tianjin, the objects of the car source city, which are beijing or tianjin, are all matched with the intermediate experimental rule a; the vehicle source city in the intermediate experiment rule B is beijing, and only the object whose vehicle source city is beijing is matched with the intermediate experiment rule B. If the city to which a certain object belongs is Beijing, the object is matched with the two intermediate experiment rules A and B, but the parameter range of the intermediate experiment rule B is smaller, so that the object is more matched with the intermediate experiment rule B, namely the matching degree of the intermediate experiment rule B is higher. In addition, the smaller the default number of items is, the smaller the parameter range is to some extent, and the default number of items may be used alternatively or both in practical application, which is not limited in this embodiment.

It should be noted that in this embodiment, "the subject parameter includes one or more of the default number of items, the creation time, and the parameter range" means that the subject parameter may directly and explicitly include the default number of items, the creation time, the parameter range, and the like, or may uniquely determine the default number of items, the creation time, and the parameter range based on the subject parameter.

Step 104: determining a first actual quantity ratio of objects corresponding to the effective experiment rules to be distributed to different models, determining a model with the minimum current ratio according to the first actual quantity ratio and a first standard quantity ratio of the effective experiment rules, and distributing target objects to the model with the minimum current ratio.

In the embodiment of the invention, if the object parameter of a certain object is matched with the realization object parameter of the effective experiment rule, the object is the object corresponding to the effective experiment rule; with the increase of the number of the acquired objects, more and more objects correspond to the effective experimental rule and are distributed to the preset model, so that the model can correspondingly process the objects. When a target object needing to be configured is obtained currently, firstly, the condition that the object is allocated to a model before is determined, namely, a first actual quantity ratio is determined; determining which proportion item has the least allocated object ratio according to the difference between the current first actual quantity ratio and a preset first standard quantity ratio, and further determining a model with the least current ratio; in this embodiment, the current target object is distributed and configured with the model with the minimum current occupation ratio, so that the first actual quantity ratio can be dynamically adjusted to ensure that the first actual quantity ratio is the same as or similar to the preset first standard quantity ratio as much as possible.

In this embodiment, the target object is distributed and configured to the model with the smallest current proportion, and the object can be distributed and configured according to the preset proportion, so that the object distribution under each experimental rule is ensured to be uniform as much as possible, the local distribution is optimal under the minimum granularity of the experimental rule, and the relative optimization of the overall distribution condition of the object can be further realized. In addition, in a short period, due to the limitation of the number, the overall distribution may be uneven; but over time, when the number of objects increases to some extent, the overall distribution will also be a more uniform state.

Optionally, the step 104 of "determining the model with the minimum current ratio according to the first actual quantity ratio and the first standard quantity ratio of the effective experimental rule" includes:

step A1: and determining a first actual quantity corresponding to each proportion item of the first actual quantity ratio, and determining a first standard configuration value corresponding to each proportion item of the first standard quantity ratio of the effective experimental rule.

In the embodiment of the invention, the first actual quantity ratio and the first standard quantity ratio are ratios of a plurality of numerical values, and each ratio item of the first actual quantity ratio and the first standard quantity ratio can correspond to a corresponding numerical value. For example, if the first actual quantitative ratio is 100:10:20, the first actual quantitative ratio includes three proportional terms, and the first actual quantitative ratios corresponding to the three proportional terms are 100, 10, and 20 in sequence. The first standard quantity ratio is similar to this, and is not described herein again.

The first actual quantity ratio may not be reduced, and each time an object is added, the corresponding proportion item is added. For example, the first actual quantitative ratio is 100:10:20, which may not be divided into about 10:1:2 in this embodiment; if a new object is assigned to the effective experimental rule and the object is assigned to the model corresponding to the intermediate proportion term, an addition process may be performed on the corresponding intermediate proportion term, and then the first actual quantitative ratio is updated to 100:11: 20.

Step A2: and determining the ratio of the first actual quantity to the corresponding first standard configuration value, and taking the model corresponding to the minimum ratio as the model with the minimum current ratio.

In the embodiment of the invention, each proportion item of the first actual quantity proportion and the first standard quantity proportion corresponds to one model; for a certain proportion item, if the ratio between the first actual quantity and the corresponding first standard configuration value is smaller, it indicates that the quantity of the objects allocated to the proportion item is smaller, and at this time, it is necessary to allocate a newly added object (i.e., a target object) to the proportion item, that is, a model corresponding to the proportion item is used as a model with the minimum current occupancy, and the current target object is allocated to the model with the minimum current occupancy. For example, the first standard quantity proportion of the effective experimental rule represents the ratio of the quantity of the objects allocated to the model a to the quantity of the objects allocated to the model B, and is 5:1, and the first standard configuration values corresponding to the two proportion terms of the first standard quantity proportion are 5 and 1 respectively; if the current first actual quantity ratio is 51:10, the ratio of the two proportion terms is 51/5 and 10/1 respectively, at the moment, the ratio of the second proportion term is the minimum, the model B corresponding to the minimum ratio is taken as the model with the minimum current ratio, the current target object is distributed to the model B, and the later first actual quantity ratio is updated to 51: 11. When a new target object is acquired again, the process of shunting the configuration object is repeated by taking 51:11 as a new first actual quantity ratio. The "distribution configuration" in this embodiment refers to configuring the objects to different models as much as possible, so as to distribute the objects, and the consistency between the first actual quantity ratio and the preset first standard quantity ratio can be ensured as much as possible, so that the first actual quantity ratio and the preset first standard quantity ratio are the same or similar.

In addition, if a plurality of minimum ratios exist currently, one ratio can be randomly selected from the plurality of minimum ratios, and the model corresponding to the selected minimum ratio is used as the model with the minimum current ratio; alternatively, a ratio may be randomly selected from a plurality of minimum ratios, and the model corresponding to the ratio is not regarded as the minimum-occupied model in the last round of allocation.

According to the method for distributing the configuration objects, the experiment rules are preset, and distribution is carried out based on the experiment rules when the target objects needing distribution configuration are obtained, so that the target objects can be distributed and configured in real time, the real-time performance is good, and the experiment range is large; meanwhile, the target object is distributed to the model with the least current occupation ratio, the object can be uniformly distributed to different models, uniform distribution is realized, and the local part and the whole part can gradually reach the optimal state. In addition, the objects are distributed to the corresponding models to be processed, the processing results are convenient to count, the processing results of different models can be compared, and the models can be conveniently trained and optimized subsequently.

On the basis of the above embodiment, the experimental rule further includes a second standard quantity matching ratio for allocating the object to different hosting states, where the hosting state in this embodiment includes at least two of a full hosting state, a half hosting state, and an unmanaged state; wherein, the fully hosting state refers to that the object is processed by the model completely without manual intervention; the semi-hosting state refers to that after the model processes the object, the processing result is continuously adjusted (or not adjusted) by the human; the unmanaged state is when the object is completely handled by a human, the model does not do any processing, or the object is simply exposed to a human worker.

At this time, after "determining an effective experimental rule corresponding to the target object in the experimental rule according to the object parameter of the target object" in step 103, the method further includes a process of configuring the object shunt to a corresponding hosting state, where the process specifically includes:

step B1: determining a second actual quantity ratio of objects corresponding to the effective experiment rules to be distributed to different hosting states, determining the hosting state with the minimum current occupation ratio according to the second actual quantity ratio and a second standard quantity ratio of the effective experiment rules, and shunting and configuring the target objects to the hosting state with the minimum current occupation ratio.

In the embodiment of the invention, the managed state can be allocated to the object, wherein the process of allocating the managed state to the object is similar to the process of allocating the model. Specifically, when a target object to be configured is currently acquired, a condition that the previous object is allocated to a different hosting state, that is, a second actual quantity ratio is determined; determining which proportion item has the least allocated object ratio according to the difference between the current second actual quantity ratio and a preset second standard quantity ratio, and further determining the hosting state with the least current ratio; in this embodiment, the current target object is distributed and configured in the hosting state with the minimum current occupation ratio, so that the second actual quantity ratio can be dynamically adjusted to ensure that the second actual quantity ratio is the same as or similar to the preset second standard quantity ratio as much as possible.

Optionally, the step B1 of "determining the current hosting state with the least percentage according to the second actual quantity ratio and the second standard quantity ratio of the effective experimental rule" includes:

step B11: and determining a second actual quantity corresponding to each proportion item of the second actual quantity ratio, and determining a second standard configuration value corresponding to each proportion item of the second standard quantity ratio of the effective experimental rule.

Step B12: and determining the ratio of the second actual quantity to the corresponding second standard configuration value, and taking the hosting state corresponding to the minimum ratio as the current hosting state with the minimum ratio.

In the embodiment of the present invention, the process of determining the hosting status with the minimum current occupancy ratio based on the ratio between the second actual quantity and the corresponding second standard configuration value is similar to the process of determining the model with the minimum current occupancy ratio based on the ratio between the first actual quantity and the corresponding first standard configuration value in the above steps a1 to a2, and both are the proportion item for determining the minimum ratio, so as to determine the corresponding hosting status (or model), which is not described in detail herein.

Optionally, different managed states may also correspond to different models, that is, if an object is assigned to a managed state, the object can only be assigned to a model corresponding to the managed state or one of some models. In this embodiment, the objects assigned to the semi-tubular state or the unmanaged state may be regarded as not being processed by the model, and therefore, the model assignment may be preferentially performed only on the objects in the fully managed state. Specifically, when the hosting state includes a complete hosting state, the step 104 "determining a first actual quantity ratio at which the object corresponding to the valid experiment rule is allocated to the different models" includes:

determining the objects which are distributed to the full hosting state, and determining a first actual quantity ratio which corresponds to the effective experiment rule and is distributed to different models.

In the embodiment of the present invention, the first standard quantity matching refers to a quantity proportion of objects in a fully managed state, which need to be allocated by different models. Specifically, when a target object needing to be allocated is acquired, firstly, which hosting state the target object needs to be allocated to is determined; if the target object is allocated to the full hosting state, determining a model with the minimum current ratio based on the first actual quantity ratio and the first standard quantity ratio, and allocating the target object to the model with the minimum current ratio. In this embodiment, the hosting state is first allocated in a shunting manner, and then the configuration model is allocated in a shunting manner, so that multi-layer shunting configuration of the target object is realized, and the shunting configuration is more accurate.

In addition, corresponding models may be preset for other managed states (i.e., a semi-managed state and an unmanaged state), and if a certain target object is assigned to the other managed states, the target object may be directly assigned to the corresponding models. Alternatively, similar to the fully hosted state, for the semi-hosted state, a corresponding standard quantity ratio (similar to the first standard quantity ratio) for assigning the model quantity may also be set, i.e. for the target objects assigned to the semi-hosted state, which model to assign specifically is also determined based on the standard quantity ratio.

On the basis of the above embodiment, after "configure target object shunting to the model with the least current proportion" in step 104, the method further includes:

step C1: and determining the current evaluation value of the target object according to the model of the shunting configuration.

Step C2: a history evaluation value of the target object is acquired, and if the current evaluation value is different from the history evaluation value, an evaluation value change message is generated.

In the embodiment of the invention, each target object can be distributed and configured to a certain model, and the model can evaluate the target object based on the attribute parameters of the target object, so as to generate a corresponding evaluation value. For example, the target object may be a vehicle source, and the model may be a vehicle sales model that determines a price of the vehicle source based on attributes of the vehicle source (e.g., age, mileage, etc., vehicle type and vehicle family), and the price may be an evaluation value of the vehicle source. In the embodiment of the present invention, the model updates the evaluation value of the object at regular time, and if the currently determined evaluation value (i.e., the current evaluation value) of the model is different from the previously determined evaluation value (i.e., the historical evaluation value), a message for changing the evaluation value of the object, i.e., an evaluation value change message, is generated to ensure timeliness of the evaluation value of the object.

Alternatively, if the current evaluation value of the target object is lower than the historical evaluation value, the evaluation value of the target object needs to be decreased, and at this time, the evaluation value of the target object may be directly updated to the current evaluation value. Alternatively, other means of reducing the value may be used, such as by issuing a coupon or giving a gift to another product. Specifically, for example, if the target is a car source and the evaluation value is a price, if the car source needs to be reduced, the price of the car source may be divided into a reduced price, an insurance price, a maintenance price, and the like, wherein the insurance price and the maintenance price may both be issued by an indirect price reduction.

According to the method for distributing the configuration objects, the experiment rules are preset, and distribution is carried out based on the experiment rules when the target objects needing distribution configuration are obtained, so that the target objects can be distributed and configured in real time, the real-time performance is good, and the experiment range is large; meanwhile, the target object is distributed to the model with the least current occupation ratio, the object can be uniformly distributed to different models, uniform distribution is realized, and the local part and the whole part can gradually reach the optimal state. In addition, the objects are distributed to the corresponding models to be processed, the processing results are convenient to count, the processing results of different models can be compared, and the models can be conveniently trained and optimized subsequently. In addition, the target object is allocated to the current hosting state with the least occupation ratio, and the refined shunting configuration of the object can be realized; and the target object is subjected to multi-layer shunting configuration, so that the shunting configuration result is more accurate. The evaluation value of the object can be updated in a timed or real-time manner through the evaluation value changing message, and the timeliness of the evaluation value of the object can be ensured.

The above describes in detail the flow of the method for shunting configuration objects, which may also be implemented by a corresponding apparatus, and the structure and function of the apparatus are described in detail below.

An apparatus for distributing objects according to an embodiment of the present invention is shown in fig. 3, and includes:

a rule setting module 31, configured to preset an experiment rule, where the experiment rule includes experiment object parameters and a first standard quantity ratio for allocating objects to different models;

an obtaining module 32, configured to obtain a target object and determine an object parameter of the target object;

a rule matching module 33, configured to determine, according to the object parameters of the target object, effective experiment rules corresponding to the target object in the experiment rules, where the effective experiment rules are experiment rules having experiment object parameters matching the object parameters of the target object;

and the distribution model module 34 is configured to determine a first actual quantity ratio at which the objects corresponding to the effective experimental rules are distributed to different models, determine a model with the minimum current ratio according to the first actual quantity ratio and a first standard quantity ratio of the effective experimental rules, and distribute the target objects to the model with the minimum current ratio.

On the basis of the foregoing embodiment, the determining, by the distribution model module 34, the model with the least current proportion according to the first actual quantity proportion and the first standard quantity proportion of the effective experimental rule includes:

determining a first actual quantity corresponding to each proportion item of the first actual quantity proportion, and determining a first standard configuration value corresponding to each proportion item of the first standard quantity proportion of the effective experimental rule;

and determining the ratio of the first actual quantity to the corresponding first standard configuration value, and taking the model corresponding to the minimum ratio as the model with the minimum current ratio.

On the basis of the above embodiment, the experimental rule further includes a second standard quantity matching ratio for allocating the object to different hosting states; the hosting state comprises at least two of a fully hosting state, a semi hosting state, and an unmanaged state;

the apparatus also includes an assign escrow status module;

after the rule matching module 32 determines, according to the object parameters of the target object, a valid experiment rule corresponding to the target object in the experiment rules, the allocation hosting status module is configured to:

determining a second actual quantity ratio of objects corresponding to the effective experiment rule to be allocated to different hosting states, determining the hosting state with the minimum current occupation ratio according to the second actual quantity ratio and a second standard quantity ratio of the effective experiment rule, and shunting and configuring the target objects to the hosting state with the minimum current occupation ratio.

On the basis of the above embodiment, the determining, by the distribution hosting state module, the hosting state with the smallest current percentage according to the second actual quantity ratio and the second standard quantity ratio of the effective experimental rule includes:

determining a second actual quantity corresponding to each proportion item of the second actual quantity ratio, and determining a second standard configuration value corresponding to each proportion item of the second standard quantity ratio of the effective experimental rule;

and determining the ratio of the second actual quantity to the corresponding second standard configuration value, and taking the hosting state corresponding to the minimum ratio as the current hosting state with the minimum ratio.

On the basis of the above embodiment, the managed state comprises a fully managed state;

the assigning model module 34 determines a first actual quantitative ratio at which objects corresponding to the valid experimental rules are assigned to different models, including:

determining all the objects which are distributed to the full hosting state, and determining a first actual quantity ratio which corresponds to the effective experiment rule and is distributed to the different models.

On the basis of the above embodiment, the determining, by the rule matching module 33, a valid experiment rule corresponding to the target object in the experiment rule according to the object parameter of the target object includes:

determining all experimental rules with experimental subject parameters matching the subject parameters of the target subject, taking the determined experimental rules as intermediate experimental rules;

determining the matching degree of each intermediate experiment rule according to the experiment object parameters of the intermediate experiment rules, and taking the intermediate experiment rule with the highest matching degree as an effective experiment rule corresponding to the target object;

the experimental object parameters comprise one or more of default item quantity, creation time and parameter range, and the lower the default item quantity of the intermediate experimental rule, the closer the creation time to the current time or the smaller the parameter range, the higher the matching degree of the intermediate experimental rule.

On the basis of the above embodiment, the apparatus further comprises a change evaluation value module;

after the assignment model module 34 allocates the target object diversion to the model with the least current occupancy, the change evaluation value module is configured to:

determining the current evaluation value of the target object according to the model of the shunting configuration; and acquiring a historical evaluation value of the target object, and generating an evaluation value change message if the current evaluation value is different from the historical evaluation value.

According to the device for distributing the configuration objects, the experiment rules are preset, and distribution is performed based on the experiment rules when the target objects needing distribution configuration are obtained, so that the target objects can be distributed and configured in real time, the real-time performance is good, and the experiment range is large; meanwhile, the target object is distributed to the model with the least current occupation ratio, the object can be uniformly distributed to different models, uniform distribution is realized, and the local part and the whole part can gradually reach the optimal state. In addition, the objects are distributed to the corresponding models to be processed, the processing results are convenient to count, the processing results of different models can be compared, and the models can be conveniently trained and optimized subsequently. In addition, the target object is allocated to the current hosting state with the least occupation ratio, and the refined shunting configuration of the object can be realized; and the target object is subjected to multi-layer shunting configuration, so that the shunting configuration result is more accurate. The evaluation value of the object can be updated in a timed or real-time manner through the evaluation value changing message, and the timeliness of the evaluation value of the object can be ensured.

Embodiments of the present invention further provide a computer storage medium, where the computer storage medium stores computer-executable instructions, which include a program for executing the method for shunting configuration objects described above, and the computer-executable instructions may execute the method in any of the method embodiments described above.

The computer storage medium may be any available medium or data storage device that can be accessed by a computer, including but not limited to magnetic memory (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.), optical memory (e.g., CD, DVD, BD, HVD, etc.), and semiconductor memory (e.g., ROM, EPROM, EEPROM, nonvolatile memory (NANDFLASH), Solid State Disk (SSD)), etc.

Fig. 4 shows a block diagram of an electronic device according to another embodiment of the present invention. The electronic device 1100 may be a host server with computing capabilities, a personal computer PC, or a portable computer or terminal that is portable, or the like. The specific embodiment of the present invention does not limit the specific implementation of the electronic device.

The electronic device 1100 includes at least one processor (processor)1110, a Communications Interface 1120, a memory 1130, and a bus 1140. The processor 1110, the communication interface 1120, and the memory 1130 communicate with each other via the bus 1140.

The communication interface 1120 is used for communicating with network elements including, for example, virtual machine management centers, shared storage, etc.

Processor 1110 is configured to execute programs. Processor 1110 may be a central processing unit CPU, or an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits configured to implement embodiments of the present invention.

The memory 1130 is used for executable instructions. The memory 1130 may comprise high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory. The memory 1130 may also be a memory array. The storage 1130 may also be partitioned and the blocks may be combined into virtual volumes according to certain rules. The instructions stored by the memory 1130 are executable by the processor 1110 to enable the processor 1110 to perform the method of shunting configuration objects in any of the method embodiments described above.

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 claims.

Claims

1. A method for streaming configuration objects, comprising:

2. The method of claim 1, wherein the determining the model with the least current proportion according to the first actual quantity ratio and the first standard quantity ratio of the effective experimental rule comprises:

3. The method of claim 1, wherein the experimenting rule further comprises a second standard quantity ratio that assigns objects to different hosting states; the hosting state comprises at least two of a fully hosting state, a semi hosting state, and an unmanaged state;

after the determining, according to the object parameters of the target object, a valid experiment rule corresponding to the target object in the experiment rules, the method further includes:

4. The method of claim 3, wherein the determining the current least populated hosting state according to the second actual quantity ratio and the second standard quantity ratio of the valid experimental rule comprises:

5. The method of claim 3, wherein the managed state comprises a fully managed state;

the determining a first actual quantitative ratio that objects corresponding to the valid experimental rules are assigned to different models comprises:

6. The method of claim 1, wherein the determining, from the object parameters of the target object, a valid experiment rule corresponding to the target object in the experiment rules comprises:

7. The method according to any one of claims 1-6, further comprising, after the configuring the target object offload to a model with a least current proportion, the steps of:

determining the current evaluation value of the target object according to the model of the shunting configuration;

and acquiring a historical evaluation value of the target object, and generating an evaluation value change message if the current evaluation value is different from the historical evaluation value.

8. An apparatus for streaming configuration objects, comprising:

9. A computer storage medium having stored thereon computer-executable instructions for performing the method of offloading a configuration object of any of claims 1-7.

10. An electronic device, comprising:

at least one processor; and the number of the first and second groups,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of shunting configuration objects of any of claims 1-7.