CN112799710B - Model release system and model release method - Google Patents
Model release system and model release method Download PDFInfo
- Publication number
- CN112799710B CN112799710B CN202110128686.2A CN202110128686A CN112799710B CN 112799710 B CN112799710 B CN 112799710B CN 202110128686 A CN202110128686 A CN 202110128686A CN 112799710 B CN112799710 B CN 112799710B
- Authority
- CN
- China
- Prior art keywords
- model
- release
- released
- service forwarding
- display device
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000013461 design Methods 0.000 claims abstract description 7
- 238000005538 encapsulation Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 abstract description 23
- 238000013508 migration Methods 0.000 abstract description 6
- 230000005012 migration Effects 0.000 abstract description 6
- 238000011161 development Methods 0.000 description 21
- 230000018109 developmental process Effects 0.000 description 21
- 238000012545 processing Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 8
- 238000012549 training Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000012546 transfer Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000004806 packaging method and process Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000008140 language development Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/70—Software maintenance or management
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The application provides a model release system and a model release method, wherein the model release system comprises a model generation device, a model service forwarding device and a model release display device, wherein the model generation device generates a model to be released according to a design strategy corresponding to a preset operation environment under the preset operation environment selected by an engineer, and sends the generated model to be released to the model service forwarding device; the model service forwarding device determines the release state corresponding to the received model to be released, and sends the determined release state corresponding to the model to be released to the model release display device, and when the model release display device receives the corresponding release state, the model to be released is displayed according to the corresponding release state, so that the migration and release time of different environments in the model generation process are shortened, the model release process is simplified, and the release efficiency of the model can be improved.
Description
Technical Field
The application relates to the technical field of model processing, in particular to a model release system and a model release method.
Background
With the development of scientific technology and the rapid development of deep learning technology, more and more technical fields select and apply trained models to develop business, thereby providing great convenience for business development.
At present, a manual release generating mode is generally adopted for releasing the models, in the process of generating the models and releasing the models, developers are required to know the models deeply, a plurality of models are generated under different development environments, the models are online to corresponding release systems, in the process of releasing the models, release of one model is required to be frequently switched under different development environments and different release systems, the release period of the models is greatly increased, and the release efficiency of the models is low.
Disclosure of Invention
In view of the above, the present application aims to provide a model publishing system and a model publishing method, which generate corresponding models to be published under different operating environments, publish and display the models to be published in a corresponding publishing state by one key, shorten the migration and publishing time of different environments in the model generating process, simplify the model publishing process, and improve the model publishing efficiency.
The embodiment of the application provides a model release system, which comprises a model generation device, a model service forwarding device and a model release display device;
the model generating device is used for generating a model to be distributed based on a preset design strategy in a preset operation environment and sending the model to be distributed to the model service forwarding device;
the model service forwarding device is used for determining the release state corresponding to the model to be released, and sending the release state corresponding to the model to be released to the model release display device;
the model release display device is used for displaying the model to be released according to the received release state.
Further, the model issuing system further comprises an encapsulation management device, wherein the encapsulation management device is used for:
receiving model release information sent by the model release display device, completing registration of a model to be released according to the model release information, and sending the model release information to the model service forwarding device so that the model service forwarding device can determine a release state corresponding to the model to be released.
Further, the model service forwarding device is further configured to:
and determining a release state corresponding to the model to be released based on the received model release information, and sending the release state to the encapsulation management device for storage.
Further, the model release display device is further used for:
when a model calling instruction is received, detecting whether a model indicated to be called in the model calling instruction is a model to be called for the first time, if so, generating a model data query instruction, sending the model data query instruction to the model service forwarding device, loading the model indicated to be called in the model calling instruction based on model operation information fed back by the model service forwarding device, and displaying the model data query instruction to a user.
Further, the model service forwarding device is further configured to:
determining identification information of a model to be queried from the received model data query instruction, retrieving model operation information corresponding to the identification information of the model to be queried from a pre-generated model database, and sending the model operation information to the model release display device.
Further, the model release display device is further used for:
and if the called model indicated in the model calling instruction is not the first called model, sending a model running information calling instruction to a packaging management device, and acquiring model running information from the model service forwarding device through the packaging management device.
Further, the model publishing system further includes a model storage device for:
and storing the model to be distributed generated by the model generating device, sending a model file pulling instruction to the model service forwarding device, and correspondingly storing the received model file data fed back by the model service forwarding device.
Further, the model release display device is further used for:
and displaying the model to be distributed in a position area corresponding to the model type according to the model type to which the model to be distributed belongs.
The embodiment of the application also provides a model release method which is applied to the model service forwarding device in the model release system, and comprises the following steps:
determining a release state corresponding to the received model to be released;
and sending the determined release state and the received model to be released to a model release display device so that the model release display device releases the model to be released according to the release state.
Further, determining the release state corresponding to the to-be-released model through the following steps:
and determining the release state corresponding to the model to be released based on the received model release information.
The embodiment of the application also provides electronic equipment, which comprises: a processor, a memory and a bus, said memory storing machine readable instructions executable by said processor, said processor and said memory communicating over the bus when the electronic device is running, said machine readable instructions when executed by said processor performing the steps of the model issuing method as described above.
The embodiment of the present application also provides a computer readable storage medium, on which a computer program is stored, which when being executed by a processor performs the steps of the model issuing method as described above.
According to the model release system and the model release method provided by the embodiment of the application, a model generation device generates a model to be released according to a design strategy corresponding to a preset operation environment under the preset operation environment selected by an engineer, and sends the generated model to be released to a model service forwarding device; the model service forwarding device determines the release state corresponding to the received model to be released, and sends the determined release state corresponding to the model to be released to the model release display device, and when the model release display device receives the corresponding release state, the model to be released is displayed according to the corresponding release state, so that the migration and release time of different environments in the model generation process are shortened, the model release process is simplified, and the release efficiency of the model can be improved.
In order to make the above objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of a model publishing system according to an embodiment of the present application;
FIG. 2 is a second schematic diagram of a model publishing system according to an embodiment of the present application;
fig. 3 is a schematic diagram of data flow in a model publishing system according to an embodiment of the present application;
FIG. 4 is a flowchart of a model publishing method according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Icon: 100-model publishing system; 110-model generating means; 120-model service forwarding means; 130-model release display device; 140-packaging management device; 150-model storage means; 160-load balancing means; 500-an electronic device; 510-a processor; 520-memory; 530-bus.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the application, as presented in the figures, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. Based on the embodiments of the present application, every other embodiment obtained by a person skilled in the art without making any inventive effort falls within the scope of protection of the present application.
First, an application scenario to which the present application is applicable will be described. The application can be applied to the technical field of model processing.
According to research, at the present stage, a manual release generation mode is generally adopted for releasing the model, in the process of generating the model and releasing the model, a developer needs to know the model deeply, a plurality of models are generated under different development environments, the models are online to corresponding release systems, in the release process of the model, release of one model needs to be frequently switched in different development environments and different release systems, the release period of the model is greatly increased, and the release efficiency of the model is low.
Based on the above, the embodiment of the application provides a model release system, which shortens the migration and release time of different environments in the model generation process, simplifies the model release process and improves the efficiency of model release.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a model publishing system 100 according to an embodiment of the present application, and as shown in fig. 1, an embodiment of the present application provides a model publishing system 100, including a model generating device 110, a model service forwarding device 120, and a model publishing display device 130. The model generating device 110 generates a model to be distributed according to a design strategy corresponding to a preset operation environment under the preset operation environment selected by an engineer, and sends the generated model to be distributed to the model service forwarding device 120; the model service forwarding device 120 determines the release state corresponding to the received to-be-released model, and sends the determined release state corresponding to the to-be-released model to the model release display device 130, and when the model release display device 130 receives the corresponding release state, the to-be-released model is displayed according to the corresponding release state.
Specifically, the model generating device 110 is configured to generate a model to be distributed based on a preset design policy in a preset operation environment, and send the model to be distributed to the model service forwarding device 120.
Here, the model to be developed is developed by a development engineer in a corresponding development environment, and the types of the model may be classified into models with different functions corresponding to different industries according to project requirements, and the industries may include: advertising industry, security industry, automotive industry, news service industry, internet industry, financial industry, etc.; the corresponding model types may be data processing class models, text analysis class models, image recognition class models, text recognition class models, predictive class models, and the like.
Here, the preset operation environments are different development language environments, in the embodiment of the present application, the development engineering is development performed under a notbook environment, and different development languages, for example, java, c++, python, etc., may be used in the development process of the model to be released, that is, the model release system 100 provided in the embodiment of the present application is compatible with multiple development languages in the model development stage.
The NoteBook is an online interactive development and debugging tool, does not need to pay attention to installation configuration, directly uses the NoteBook to write and debug model training codes, and then carries out model training and model deployment based on the codes without multi-platform switching.
Wherein, the NoteBook has the following characteristics: the interactive code development is supported, the code is quickly created, edited and operated, and the chart is displayed; model development training is carried out, multiple languages are supported, and environment switching is not needed; the installation of an open source framework is supported, and the interconversion of a network structure and codes is supported; and the CPU/GPU hybrid scheduling is supported, and the resource multiplexing is efficient.
The running flow of the NoteBook architecture may be: the user enters a Notebook client through a browser; selecting a kernel in a Notebook client; in a notbook, sample data may be uploaded; in a Notebook, a model algorithm is built, and model training is carried out; the model after training is found in a Notebook, and the generated model can be subjected to one-key API and SDK; each service system can call an API or use an SDK to issue the AI intelligent application.
Here, the model generating apparatus 110 supports model training of a single machine or a cluster in the model development process, supports various language development models, and does not need to switch environments.
Further, the model service forwarding device 120 is configured to determine a release state corresponding to the to-be-released model, and send the release state corresponding to the to-be-released model to the model release display device 130.
In this embodiment of the present application, the release status of the model to be released is typically shown in the form of an interface or a data packet, so that a developer needing to call the related model calls the interface or the data packet of the corresponding model when deploying the process by himself.
Here, the model service forwarding device 120 (i.e., model service) binds the release state with the corresponding model after determining the release state of the model to be released, so as to release the model to be released in the corresponding release state.
Here, the manner of binding the release state with the corresponding model may be to determine the identification information (number, name, etc.) of the model to be released as the naming of the corresponding release state; or storing the corresponding relation between the release state and the corresponding model in a database, and determining the corresponding relation between the release state and the model from the database when needed.
Here, the functions of the model service forwarding apparatus 120 include multi-version model management, guaranteed logical consistency, zero delay of model on-line, deletion of screening indexes, and the like, in addition to determining the release state of the model to be released.
The model service forwarding device 120 is also compatible with the deployment of multiple languages such as Java, pyhton, C ++, and has high flexibility.
Here, each model to be distributed is deployed with one micro-service in the model service forwarding apparatus 120, and different models are managed based on different micro-services.
Further, the model publishing display device 130 is configured to display the model to be published according to the received publishing status.
Here, the model release display device 130 includes a multi-industry, multi-function, high-efficiency model market, and whether there is a mature model that is suitable for can be directly searched in the model release display device 130, so that the model achievement at the forefront can be directly enjoyed, and meanwhile, multi-version iteration of the model is supported in the model release display device 130.
Here, the model release presentation apparatus 130 includes a display interface and a usage instruction component for sdk loading model calls, and together completes release display of the model to be released.
The model publishing and exposing device 130 displays the model to be published on the front-end interface, typically through a Rest API.
Here, the model release presentation apparatus 130 may support three different model release modes: and after the NoteBook trains the model, the direct page clicks are released, and after the hao model module trains the model, the direct page clicks are released, and a third party directly releases the direct page clicks through an releasing model API.
Further, referring to fig. 2, fig. 2 is a second schematic structural diagram of the model publishing system 100 according to the embodiment of the present application, as shown in fig. 2, the model publishing system 100 further includes a package management device 140, where the package management device 140 is configured to:
receiving the model release information sent by the model release display device 130, completing the registration of the model to be released according to the model release information, and sending the model release information to the model service forwarding device 120; for the model service forwarding device 120 to determine the publishing status corresponding to the model to be published.
Here, the model distribution presentation apparatus 130 registers the model distribution information in the package management apparatus 140, and completes the registration of the model distribution information and the determination of the distribution rule of the model distribution information.
The encapsulation management device 140 may be a Zookeeper cluster, where Zookeeper is an efficient distributed coordination service, and may provide services such as configuration information management, naming, distributed synchronization, cluster management, database switching, etc. It is not suitable for storing large amounts of information and can be used to store small amounts of information such as configurations, publications and subscriptions. The number of nodes in the Zookeeper cluster is generally odd number (> =3), if the Master in the cluster is hung up, and when the number of the rest nodes is more than half, the new Master node can be lifted, and the service can be continuously provided for the outside.
Here, the model service forwarding apparatus 120 may access the encapsulation management apparatus 140 through a heartbeat mechanism, and only when the model service forwarding apparatus 120 accesses the encapsulation management apparatus 140, the encapsulation management apparatus 140 may send the model issue information to the model service forwarding apparatus 120, so that the model service forwarding apparatus 120 determines an issue state corresponding to the model to be issued.
Further, as shown in fig. 2, the model publishing system 100 further includes a model storage device 150, where the model storage device 150 is configured to:
the model to be distributed generated by the model generating device 110 is stored, and a model file pulling instruction is sent to the model service forwarding device 120, so that the received model file data fed back by the model service forwarding device 120 is correspondingly stored.
Here, the model generating apparatus 110 performs model training based on the build model algorithm by using the sample data uploaded by the notbook, and stores the model to be distributed in the model storage apparatus 150 after obtaining the model to be distributed.
When the model to be published is stored, the classification type, the model identifier, and the model version number of the model to be published need to be stored together in the model storage device 150.
Here, the model file pulling instruction sent by the model storage device 150 to the model service forwarding device 120 may include model data of at least one model history version in front of the latest stored model to be published, and the plurality of model history versions are stored together to represent the process of updating the model and the optimization part of the model that is easier to compare.
Further, the model publishing system 100 further includes a load balancing device 160, where the load balancing device 160 is configured to: the task processing amount of each service forwarding module is adjusted according to the task processing amounts of the plurality of service forwarding modules included in the model service forwarding device 120.
Here, when the model service transfer device 120 performs task processing, a plurality of service transfer modules are included in the model service transfer device 120 to perform common processing, and adjustment is required by the load balancing device 160 to balance the task processing amount of each service transfer module.
Here, when any service forwarding module included in the model service forwarding apparatus 120 is abnormal in handling, the load balancing apparatus 160 may adjust other service forwarding modules that are normally operated, so as to ensure the normal operation of the model service forwarding apparatus 120.
Further, the model service forwarding apparatus 120 is further configured to:
and determining a release state corresponding to the model to be released based on the received model release information, and sending the release state to the package management device storage 140.
Here, after receiving the model release information, the model service forwarding device 120 determines a release state of the model to be released from the model release information, converts the model to be released according to the corresponding release state, and sends the determined release state corresponding to the model to be released to the package management device storage 140 for corresponding storage.
Here, the model service forwarding device 120 generally stores the form of converting the model into a Rest API in the package management device storage 140.
Further, the model release display device 130 is further configured to:
when a model calling instruction is received, detecting whether a called model indicated in the model calling instruction is a first called model, if so, generating a model data query instruction, sending the model data query instruction to the model service forwarding device 120, loading the model indicated to be called in the model calling instruction based on model operation information fed back by the model service forwarding device 120, and displaying the model data query instruction to a user.
Here, when the model calling instruction is received, it is detected whether the model required to be called in the model calling instruction is the first called model in the model issuing system 100, that is, whether the called model is the newly issued model, if it is determined that the called model is the first called model, a model data query instruction is generated and sent to the model service forwarding device 120, and according to the model running information fed back by the model service forwarding device 120, the model indicating the called model in the model calling instruction is loaded and displayed to the user.
Here, when the model call instruction is received, according to the identification information (model number, model name, etc.) of the model required to be called in the model call instruction, searching whether a call record matched with the identification information exists in the history call information, and if the call record matched with the identification information is not searched in the history call information, determining that the model is called for the first time.
Here, the model operation information fed back by the model service forwarding device 120 includes the data flow in which the model is loaded, the authority of the person who can load the model, the operation language environment of the model, and the like.
The authority of the person loading the model may specify the authority of the person capable of calling the model, including the identification information of the person, etc., which may be used by the model issuing and displaying device 130 to verify the person calling the model, if the person issuing the model calling instruction does not have the authority of calling the model, the person is not like the person to display the corresponding model.
The authentication of the authority of the person may be performed by the person performing the model call or verifying the password input by the login model issuing system 100, and when the password is input, if the input password is compared with the password required to be called or preset by the login model issuing system 100, if the input password is consistent with the preset password, the person is determined to have the authority of the model call or login model issuing system 100.
Further, the model service forwarding apparatus 120 is further configured to:
determining identification information of a model to be queried from the received model data query instruction, retrieving model operation information corresponding to the identification information of the model to be queried from a pre-generated model database, and sending the model operation information to the model release display device.
Here, after receiving the model data query instruction, the model service forwarding device 120 determines the identification information of the model to be queried from the model data query instruction, retrieves the model operation information corresponding to the identification information of the model to be queried from the model database generated in advance, and then sends the determined model operation information to the model release display device 130.
Here, when the model generating device 110 generates the retransmission to the model service forwarding device 120, each model to be transmitted has corresponding model operation information, after the model service forwarding device 120 transmits the model to be transmitted to the model transmission display device 130 in a corresponding transmission state, the model operation information of each model to be transmitted is stored in a targeted manner, different model operation information is identified by the identification information of the corresponding model to be transmitted, and after receiving the model data query instruction, the model operation information is called in a targeted manner and fed back.
Further, the model release display device 130 is further configured to:
if the model indicated to be called in the model calling instruction is not the first called model, sending a model running information calling instruction to the encapsulation management device 140, and obtaining the model running information from the model service forwarding device 120 through the encapsulation management device 140.
Here, if it is determined that the model indicated to be invoked in the model invoking instruction is not the first invoked model, the model operation information invoking instruction needs to be sent to the encapsulation management device 140, and the encapsulation management device 140 indirectly obtains the model operation information from the model service forwarding device 120.
Here, when the model call instruction is received, according to the identification information (model number, model name, etc.) of the model required to be called in the model call instruction, searching whether a call record matched with the identification information exists in the history call information, and if the call record matched with the identification information exists in the history call information, determining that the model is not called for the first time.
Further, the model release display device 130 is further configured to:
and displaying the model to be distributed in a position area corresponding to the model type according to the model type to which the model to be distributed belongs.
Here, for different model types, different display areas may be corresponding to each other in the interface during display, and it is necessary to determine a display position area of the model to be released according to the different model types, and display the model to be released in the corresponding position area.
Here, the model release presentation apparatus 130 may further include a model list, model classification, and popularity.
Wherein the model list represents: the model list shows a list of all models in the model release presentation 130, and the model list deployment status is divided into enable (available) and disable (unavailable). Clicking the download in the operation, and downloading the model file; clicking on the model file name in the "model list" can look at model details, which include: model name, model ID, model version, model description, etc. Clicking on the "blue dot" next to the "model sdk download" button may view sdk the instructions for loading the model call.
The step of sdk loading may be: downloading a model file with a model name of 'test_1'; clicking on the "model sdk download" button; the model "test_1" file downloaded locally and the sdk file downloaded locally are integrated into the project.
Wherein the model classification represents: model classification pages are currently classified according to the industry of models, model types. The industry includes: advertising, security, automotive, news services, internet, finance; model types include: data processing, text analysis, image recognition, text recognition, prediction.
Wherein popularity represents: popularity pages mainly show: the model published by XX (author) is downloaded YYY times and is a statistics page.
Here, the model publishing system 100 provided by the present application also supports a model sharing function, and after a development engineer develops a new model, the model may be shared by other members through the model publishing system 100.
Further, referring to fig. 3, fig. 3 is a schematic diagram of data flow in the model publishing system 100 according to the embodiment of the present application, as shown in fig. 3, the model trained by the model generating device 110 is uploaded to the model storing device 150; the model is released to the model release display device 130 in the form of a Rest API through a model release function; in the process of model release, the model release display device 130 registers the model release information to the package management device 140; the model service forwarding device 120 accesses the encapsulation management device 140 through a heartbeat mechanism to acquire model release information; the model service forwarding device 120 performs load balancing processing through the load balancing device 160; the model service forwarding apparatus 120 processes the release information and converts the model into a Rest API to store in the package management apparatus 140; model release presentation device 130 also accesses encapsulation management device 140 via a heartbeat mechanism, obtains the Rest API, and presents the Rest API at the front-end interface; and the engineer acquires the RestAPI address on the front-end page and integrates the RestAPI address into a required service system.
According to the model release system provided by the embodiment of the application, a model generation device generates a model to be released according to a design strategy corresponding to a preset operation environment under the preset operation environment selected by an engineer, and sends the generated model to be released to a model service forwarding device; the model service forwarding device determines the release state corresponding to the received model to be released, and sends the determined release state corresponding to the model to be released to the model release display device, and when the model release display device receives the corresponding release state, the model to be released is displayed according to the corresponding release state, so that the migration and release time of different environments in the model generation process are shortened, the model release process is simplified, and the release efficiency of the model can be improved.
Referring to fig. 4, fig. 4 is a flowchart of a model publishing method according to an embodiment of the present application. As shown in fig. 4, the method for publishing a model provided by the embodiment of the present application includes:
s401, determining the release state corresponding to the received to-be-released model.
In this embodiment of the present application, the release status of the model to be released is typically shown in the form of an interface or a data packet, so that a developer needing to call the related model calls the interface or the data packet of the corresponding model when deploying the process by himself.
S402, the determined release state and the received model to be released are sent to a model release display device, so that the model release display device releases the model to be released according to the release state.
Here, the model service forwarding device binds the release state with the corresponding model after determining the release state of the model to be released, so as to release the model to be released in the corresponding release state.
Here, the manner of binding the release state with the corresponding model may be to determine the identification information (number, name, etc.) of the model to be released as the naming of the corresponding release state; or storing the corresponding relation between the release state and the corresponding model in a database, and determining the corresponding relation between the release state and the model from the database when needed.
Further, determining the release state corresponding to the to-be-released model through the following steps: and determining the release state corresponding to the model to be released based on the received model release information.
In the step, after model release information is received, the release state of the model to be released is determined from the model release information, the model to be released is converted according to the corresponding release state, and the determined release state corresponding to the model to be released is converted.
Further, after step S402, the model publishing method further includes: determining identification information of a model to be queried from the received model data query instruction; and retrieving model operation information corresponding to the identification information of the model to be queried from a pre-generated model database.
Here, after receiving the model data query instruction, determining identification information of the model to be queried from the model data query instruction, and retrieving model operation information corresponding to the identification information of the model to be queried from a pre-generated model database.
Here, each model to be distributed is provided with corresponding model operation information, after the model to be distributed is received, the model operation information of each model to be distributed is stored in a targeted manner, different model operation information is identified by the identification information of the corresponding model to be distributed, and after a model data query instruction is received, the model operation information is called and fed back in a targeted manner.
The model issuing method provided by the embodiment of the application determines the issuing state corresponding to the received model to be issued; and sending the determined release state and the received model to be released to a model release display device so that the model release display device releases the model to be released according to the release state.
In this way, according to the release state of the received generation release model, the model to be released is released and displayed by one key according to the corresponding release state, so that the migration and release time of different environments in the model generation process are shortened, the model release process is simplified, and the efficiency of model release can be improved.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the application. As shown in fig. 5, the electronic device 500 includes a processor 510, a memory 520, and a bus 530.
The memory 520 stores machine-readable instructions executable by the processor 510, and when the electronic device 500 is running, the processor 510 communicates with the memory 520 through the bus 530, and when the machine-readable instructions are executed by the processor 510, the steps of the model issuing method in the method embodiment shown in fig. 4 can be executed, and the specific implementation can be referred to the method embodiment and will not be described herein.
The embodiment of the present application further provides a computer readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the steps of the model publishing method in the method embodiment shown in fig. 4 may be executed, and a specific implementation manner may refer to the method embodiment and will not be described herein.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Finally, it should be noted that: the above examples are only specific embodiments of the present application, and are not intended to limit the scope of the present application, but it should be understood by those skilled in the art that the present application is not limited thereto, and that the present application is described in detail with reference to the foregoing examples: any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or perform equivalent substitution of some of the technical features, while remaining within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application. Therefore, the protection scope of the application is subject to the protection scope of the claims.
Claims (7)
1. The model release system is characterized by comprising a model generation device, a model service forwarding device and a model release display device;
the model generating device is used for generating a model to be distributed based on a preset design strategy in a preset operation environment and sending the model to be distributed to the model service forwarding device;
the model service forwarding device is used for determining the release state corresponding to the model to be released and sending the release state corresponding to the model to be released to the model release display device;
the model release display device is used for displaying the model to be released according to the received release state;
the model release display device is also used for:
when a model calling instruction is received, detecting whether a model indicated to be called in the model calling instruction is a first called model, if so, generating a model data query instruction, sending the model data query instruction to the model service forwarding device, loading the model indicated to be called in the model calling instruction based on model operation information fed back by the model service forwarding device, and displaying the model data query instruction to a user;
wherein the model operation information includes at least one of: the data flow of the model loaded, the authority of the personnel loading the model and the operation language environment of the model;
the model issuing system further comprises an encapsulation management device, wherein the encapsulation management device is used for:
receiving model release information sent by the model release display device, completing registration of a model to be released according to the model release information, and sending the model release information to the model service forwarding device so that the model service forwarding device can determine a release state corresponding to the model to be released;
the model release display device is also used for:
and if the called model indicated in the model calling instruction is not the first called model, sending a model running information calling instruction to the encapsulation management device, and acquiring the model running information from the model service forwarding device through the encapsulation management device.
2. The model distribution system according to claim 1, wherein the model service forwarding means is further configured to:
and determining a release state corresponding to the model to be released based on the received model release information, and sending the release state to the encapsulation management device for storage.
3. The model distribution system according to claim 1, wherein the model service forwarding means is further configured to:
determining identification information of a model to be queried from the received model data query instruction, calling model operation information corresponding to the identification information of the model to be queried from a pre-generated model database, and sending the model operation information to the model release display device.
4. The model publication system of claim 1, further comprising a model storage device for:
storing the model to be distributed generated by the model generating device, and sending a model file pulling instruction to the model service forwarding device, and correspondingly storing the received model file data fed back by the model service forwarding device; the model file pulling instruction comprises model data of at least one model history version of a preamble consistent with a newly stored model to be distributed; the model file data comprises a classification type, a model identifier and a model version number of a model to be distributed.
5. The model publication system of claim 1, wherein said model publication presentation means is further for:
and displaying the model to be distributed in a position area corresponding to the model type according to the model type to which the model to be distributed belongs.
6. A model distribution method, characterized in that it is applied to a model service forwarding apparatus in a model distribution system according to any one of claims 1 to 5, the model distribution method comprising:
determining a release state corresponding to the received model to be released;
and sending the determined release state and the received model to be released to a model release display device so that the model release display device releases the model to be released according to the release state.
7. The model publishing method of claim 6, wherein the publishing state corresponding to the model to be published is determined by:
and determining the release state corresponding to the model to be released based on the received model release information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110128686.2A CN112799710B (en) | 2021-01-29 | 2021-01-29 | Model release system and model release method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110128686.2A CN112799710B (en) | 2021-01-29 | 2021-01-29 | Model release system and model release method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112799710A CN112799710A (en) | 2021-05-14 |
| CN112799710B true CN112799710B (en) | 2023-11-03 |
Family
ID=75813022
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110128686.2A Active CN112799710B (en) | 2021-01-29 | 2021-01-29 | Model release system and model release method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112799710B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764808A (en) * | 2018-03-29 | 2018-11-06 | 北京九章云极科技有限公司 | Data Analysis Services system and its on-time model dispositions method |
| CN109271602A (en) * | 2018-09-05 | 2019-01-25 | 腾讯科技（深圳）有限公司 | Deep learning model dissemination method and device |
| CN109558162A (en) * | 2017-09-26 | 2019-04-02 | 腾讯科技（深圳）有限公司 | Application issued method, apparatus, computer equipment and storage medium |
| CN110362337A (en) * | 2019-07-12 | 2019-10-22 | 腾讯科技（北京）有限公司 | Version dissemination method, device, equipment and the storage medium of application program |
| CN110427214A (en) * | 2019-07-29 | 2019-11-08 | 苏州亿歌网络科技有限公司 | A kind of dissemination method of application program, device, equipment and storage medium |
| CN111580926A (en) * | 2020-03-25 | 2020-08-25 | 中国平安财产保险股份有限公司 | Model publishing method, model deploying method, model publishing device, model deploying device, model publishing equipment and storage medium |
| CN112130865A (en) * | 2020-09-30 | 2020-12-25 | 北京明略昭辉科技有限公司 | Model management method and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9465880B2 (en) * | 2013-05-14 | 2016-10-11 | International Business Machines Corporation | Optimizing storage in a publish / subscribe environment |
| US20160232256A1 (en) * | 2015-02-06 | 2016-08-11 | Autodesk, Inc. | Model builder |
-
2021
- 2021-01-29 CN CN202110128686.2A patent/CN112799710B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109558162A (en) * | 2017-09-26 | 2019-04-02 | 腾讯科技（深圳）有限公司 | Application issued method, apparatus, computer equipment and storage medium |
| CN108764808A (en) * | 2018-03-29 | 2018-11-06 | 北京九章云极科技有限公司 | Data Analysis Services system and its on-time model dispositions method |
| CN109271602A (en) * | 2018-09-05 | 2019-01-25 | 腾讯科技（深圳）有限公司 | Deep learning model dissemination method and device |
| CN110362337A (en) * | 2019-07-12 | 2019-10-22 | 腾讯科技（北京）有限公司 | Version dissemination method, device, equipment and the storage medium of application program |
| CN110427214A (en) * | 2019-07-29 | 2019-11-08 | 苏州亿歌网络科技有限公司 | A kind of dissemination method of application program, device, equipment and storage medium |
| CN111580926A (en) * | 2020-03-25 | 2020-08-25 | 中国平安财产保险股份有限公司 | Model publishing method, model deploying method, model publishing device, model deploying device, model publishing equipment and storage medium |
| CN112130865A (en) * | 2020-09-30 | 2020-12-25 | 北京明略昭辉科技有限公司 | Model management method and system |
Non-Patent Citations (2)
| Title |
|---|
| 基于 Skyline 的海量三维模型网络发布显示;王子启等;《地理空间信息》;第12卷(第5期);全文 * |
| 基于Revit的BIM模型 质量检查与发布平台研究;操锋等;《铁路技术创新》(第4期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112799710A (en) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9239709B2 (en) | Method and system for an interface certification and design tool | |
| US20180181282A1 (en) | Providing application and device management using entitlements | |
| US8832181B2 (en) | Development and deployment of mobile and desktop applications within a flexible markup-based distributed architecture | |
| US8245129B2 (en) | Method and system for providing synchronization of directory data | |
| US7543299B2 (en) | Creating web services programs from other web services programs | |
| EP2120195A1 (en) | Client-side rule engine for executing business rules in rich internet applications | |
| US20120110093A1 (en) | Service oriented architecture version and dependency control | |
| CN108614776A (en) | Across the dynamic fixing method and system of language protocol test request | |
| CN102597993A (en) | Managing application state information by means of a uniform resource identifier (uri) | |
| CN111708702A (en) | Simulation test method, client, server, system and readable storage medium | |
| US20210081263A1 (en) | System for offline object based storage and mocking of rest responses | |
| US7996840B2 (en) | Method, system, and apparatus for scheduling pattern based web services | |
| US20220334959A1 (en) | Method and apparatus for generating software test reports | |
| US20040268312A1 (en) | Application development support, component invocation monitoring, and data processing | |
| CN110011875B (en) | Dial testing method, device, equipment and computer readable storage medium | |
| CN109104368B (en) | Connection request method, device, server and computer readable storage medium | |
| CN110532182B (en) | Automatic testing method and device for virtualization platform | |
| CN112799710B (en) | Model release system and model release method | |
| WO2023151397A1 (en) | Application program deployment method and apparatus, device, and medium | |
| CN115811481A (en) | Interactive service testing method and device, computer equipment and storage medium | |
| CN115291928A (en) | Task automatic integration method and device of multiple technology stacks and electronic equipment | |
| CN111400623B (en) | Method and device for searching information | |
| JP5299625B2 (en) | Operation support apparatus, operation support method, and program | |
| CN111610990B (en) | Method, device and related system for upgrading application program | |
| US20090222471A1 (en) | Database exploration for building wireless component applications |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |