Model delivery method, model delivery platform, device, and storage medium

Abstract

The application provides a model delivery method, a model delivery platform, equipment and a storage medium. The model delivery method comprises the following steps: obtaining model meta information; analyzing the model meta information to obtain training data; training a machine learning model based on the training data to obtain a trained target machine learning model; analyzing the model meta information to obtain a model environment corresponding to the machine learning model; deploying the target machine learning model based on the model environment; and enabling the target machine learning model when the deployment of the target machine learning model is completed. In the embodiment of the application, the training data is obtained through the model meta information in the model training process, and the model environment is obtained through the model meta information in the model deployment process. Therefore, the model training and the model deployment can be implemented without cross-platform, and the efficiency of model delivery is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of machine learning, and particularly relates to a model delivery method, a model delivery platform, equipment and a storage medium. BACKGROUND

[0002] The general process of AI model service from demand submission to model development and training, and then to online delivery is as follows: a data processor performs data processing on training data, an algorithm engineer develops a machine learning model and trains the machine learning model using the training data, after the machine learning model is trained, an operation and maintenance engineer deploys an environment and deploys the trained machine learning model, and then performs online inference through the machine learning model to realize model delivery.

[0003] Therefore, the development and training process of the machine learning model is completed by the algorithm engineer, and the deployment process of the machine learning model is completed by the operation and maintenance engineer. Since the algorithm engineer and the operation and maintenance engineer work through their respective development platforms, the model delivery needs to be operated across multiple platforms, which leads to low efficiency of the model delivery. SUMMARY

[0004] The main purpose of the present application is to provide a model delivery method, a model delivery platform, equipment and a storage medium, which aims to solve the technical problem that the model delivery needs to be operated across multiple platforms, which leads to low efficiency of the model delivery.

[0005] To achieve the above purpose, the present application provides a model delivery method, comprising:

[0006] obtaining model meta information;

[0007] parsing the model meta information to obtain training data;

[0008] training a machine learning model based on the training data to obtain a trained target machine learning model;

[0009] parsing the model meta information to obtain a model environment corresponding to the machine learning model;

[0010] deploying the target machine learning model based on the model environment;

[0011] in the case where the target machine learning model is deployed, enabling the target machine learning model.

[0012] Optionally, before the model meta information is obtained, the method comprises:

[0013] importing a data set; the data set comprises a plurality of candidate data;

[0014] performing data labeling on each candidate data;

[0015] delete the abnormal data in the plurality of candidate data to obtain training data; the abnormal data corresponds to a data label representing candidate data abnormality;

[0016] store the training data to model meta information.

[0017] Optionally, before the model meta information is obtained, the method comprises:

[0018] obtain model development code;

[0019] compile the model development code to generate a machine learning model;

[0020] obtain model environment code;

[0021] debug a training environment corresponding to the machine learning model based on the model environment code;

[0022] store the model development code and the model environment code to model meta information;

[0023] The model development code is used to compile the machine learning model, and the model environment code is used to compile a model environment corresponding to the machine learning model.

[0024] Optionally, the model environment is used to deploy the target machine learning model, comprising:

[0025] receive a computing power adjustment instruction and a capacity adjustment instruction;

[0026] In response to the computing power adjustment instruction, re-allocate computing power resources corresponding to the target machine learning model in a target cluster; the target cluster is a cluster where the target machine learning model is deployed;

[0027] In response to the capacity adjustment instruction, re-allocate storage resources corresponding to the target machine learning model in the target cluster;

[0028] Deploy the target machine learning model with re-allocated computing power resources and storage resources in the model environment.

[0029] Optionally, the method further comprises:

[0030] receive an account login instruction; the account login instruction carries an account name and an account password;

[0031] In response to the account login instruction, verify the account name and the account password;

[0032] If the verification is passed and a read instruction is received, read the model meta information associated with the account name;

[0033] The model meta-information includes training data, model development code, and model environment code.

[0034] To achieve the above object, the application further provides a model delivery platform, comprising a model training module, a model deployment module, and a model inference module.

[0035] The model training module is configured to obtain model meta-information.

[0036] The model meta-information is parsed to obtain training data.

[0037] The machine learning model is trained based on the training data to obtain a target machine learning model.

[0038] The model deployment module is configured to parse the model meta-information to obtain a model environment corresponding to the machine learning model.

[0039] The target machine learning model is deployed based on the model environment.

[0040] The model inference module is configured to enable the target machine learning model when the target machine learning model is deployed.

[0041] Optionally, the model delivery platform further comprises:

[0042] A data processing module is configured to import a data set, wherein the data set comprises a plurality of candidate data.

[0043] Each candidate data is labeled.

[0044] Abnormal data in the plurality of candidate data is deleted to obtain training data, wherein the abnormal data corresponds to data labels representing candidate data abnormalities.

[0045] The training data is stored in the model meta-information.

[0046] Optionally, the model delivery platform further comprises:

[0047] A model development module is configured to obtain model development code.

[0048] The model development code is compiled to generate a machine learning model.

[0049] Model environment code is obtained.

[0050] A training environment corresponding to the machine learning model is debugged based on the model environment code.

[0051] store the model development code and the model environment code to model meta-information;

[0052] The model development code is used for compiling the machine learning model, and the model environment code is used for compiling a model environment corresponding to the machine learning model.

[0053] To solve the above technical problems, the embodiment of the present application also provides a computer device, which adopts the technical scheme as follows:

[0054] The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of the model delivery method according to any one of the embodiments of the present application when executing the computer program.

[0055] To solve the above technical problems, the embodiment of the present application also provides a computer readable storage medium, which adopts the technical scheme as follows:

[0056] The computer readable storage medium stores a computer program, and the computer program implements the steps of the model delivery method according to any one of the embodiments of the present application when executed by a processor.

[0057] Compared with the prior art, the embodiment of the present application has the following beneficial effects:

[0058] The present application provides a model delivery method, a model delivery platform, a device and a storage medium. The model delivery method comprises: obtaining model meta-information; analyzing the model meta-information to obtain training data; training a machine learning model based on the training data to obtain a target machine learning model after training; analyzing the model meta-information to obtain a model environment corresponding to the machine learning model; deploying the target machine learning model based on the model environment; and enabling the target machine learning model in the case of completing the deployment of the target machine learning model. In the embodiment of the present application, the training data is obtained through the model meta-information in the model training process, and the model environment is obtained through the model meta-information in the model deployment process. Therefore, the model training and the model deployment can be realized without cross-platform, and the efficiency of model delivery is improved. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the schemes in the present application, the drawings needed in the description of the embodiments of the present application will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0060] Figure 1 is an exemplary system architecture diagram to which the present application can be applied;

[0061] Figure 2 is a flowchart of a model delivery method provided by an embodiment of the present application;

[0062] Figure 3 is a schematic diagram of an application scenario of a model delivery method provided by an embodiment of the present application;

[0063] Figure 4 is a structural schematic diagram of a model delivery platform provided by an embodiment of the present application;

[0064] Figure 5 is a structural schematic diagram of a model delivery platform provided by an embodiment of the present application;

[0065] Figure 6 is a basic structural block diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0066] The model delivery method provided by the embodiments of the present application is applied to a model delivery platform. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terms used in the specification of the application are only for the purpose of describing specific embodiments and are not intended to limit the application; the terms "include" and "have" and any variations thereof in the specification and claims of the application and the above description of drawings are intended to cover non-exclusive inclusion. The terms "first", "second" and the like in the specification and claims of the application or the above description of drawings are used to distinguish different objects, not to describe a particular order.

[0067] In this paper, the phrase "embodiment" means that the specific features, structures or characteristics described in conjunction with the embodiment can be included in at least one embodiment of the present application. The phrase appears in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment to other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0068] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings.

[0069] As shown in Figure 1 , the system architecture 100 can include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a communication link medium between the terminal devices 101, 102, 103 and the server 105. The network 104 can include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0070] The user can use the terminal device 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social online platform software, etc.

[0071] The terminal device 101, 102, 103 can be various electronic devices with a display screen and supporting web browsing, including but not limited to smartphones, tablet computers, e-book readers, MP3 (Moving Picture Experts Group Audio Layer III) players, MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, etc.

[0072] The server 105 can be a server providing various services, such as a background server supporting the pages displayed on the terminal device 101, 102, 103.

[0073] It should be noted that the model delivery method provided by the embodiments of the present application is generally executed by a server / terminal device, and accordingly, the model delivery platform is generally provided in the server / terminal device.

[0074] It should be understood that Figure 1 The number of terminal devices, networks and servers in the above description is only illustrative. According to the needs of implementation, there can be any number of terminal devices, networks and servers.

[0075] Please refer to Figure 2 , a flowchart of one embodiment of the model delivery method according to the present application is shown. The embodiments of the present application can acquire and process related data based on artificial intelligence technology.

[0076] The model delivery method provided by the embodiments of the present application is applied to a model delivery platform, which includes a data processing module, a model development module, a model training module, a model deployment module and a model inference module. Optionally, the above-mentioned model delivery platform is a machine learning platform. The model delivery method includes the following steps:

[0077] S210, acquiring model meta information.

[0078] S220, analyzing the model meta information to obtain training data.

[0079] It should be noted that the model delivery platform pre-stores model meta information, and the model meta information can be stored in a database of the model delivery platform, or the model delivery platform can obtain the model meta information through a blockchain technology.

[0080] In this step, after obtaining the model meta information, the model meta information is parsed to obtain training data.

[0081] S230, training the machine learning model based on the training data to obtain a trained target machine learning model.

[0082] In this step, after obtaining the training data, the training data can be used to train the machine learning model, and then a trained machine learning model is obtained. The trained machine learning model is referred to as a target machine learning model.

[0083] Optionally, the model meta information further includes model development code. In this step, the model meta information is parsed to obtain the model development code, and the model development code is compiled to obtain the machine learning model.

[0084] S240, parsing the model meta information to obtain a model environment corresponding to the machine learning model.

[0085] S250, deploying the target machine learning model based on the model environment.

[0086] In this step, the model meta information is parsed to obtain a model environment corresponding to the machine learning model, and the target machine learning model is deployed in the model environment.

[0087] Optionally, the model meta information further includes model environment code. In this step, the model meta information is parsed to obtain the model environment code, and the model environment code is compiled to obtain the model environment.

[0088] S260, in the case where the target machine learning model is deployed, the target machine learning model is enabled.

[0089] In this step, in the case where the target machine learning model is deployed, the target machine learning model can be put online, and the target machine learning model is enabled. Optionally, the target machine learning model can be applied to infer received service information.

[0090] In the embodiment of the present application, model meta information is acquired, training data is obtained by analyzing the model meta information, the machine learning model is trained based on the training data to obtain a trained target machine learning model, the model environment corresponding to the machine learning model is obtained by analyzing the model meta information, the target machine learning model is deployed based on the model environment, and the target machine learning model is enabled in the case that the deployment of the target machine learning model is completed. In the embodiment of the present application, the training data is acquired through the model meta information in the model training process, and the model environment is acquired through the model meta information in the model deployment process, so that the model training and the model deployment can be implemented without cross-platform, and the efficiency of model delivery is improved.

[0091] Optionally, before the model meta information is acquired, the method comprises:

[0092] The data set is imported, and the data set comprises a plurality of candidate data;

[0093] Each candidate data is data-labeled;

[0094] Abnormal data in the plurality of candidate data is deleted to obtain training data, and the data label corresponding to the abnormal data represents that the candidate data is abnormal;

[0095] The training data is stored in the model meta information.

[0096] As described above, the model delivery platform comprises a data processing module.

[0097] In the embodiment, the data processing module imports a data set in response to a data processing request, wherein the data set can be data stored locally by the machine learning platform, data acquired by the machine learning platform through the Internet, or data acquired by the machine learning platform through communication with other devices.

[0098] The data processing module data-labels each candidate data in the data set in response to a data labeling instruction.

[0099] An optional implementation is that the candidate data is manually data-labeled, and in this implementation, the data processing module sends a plurality of candidate data included in the data set to a front-end page, a data processing teacher sends a data labeling instruction to the data processing module based on the candidate data, and manual data labeling of the plurality of candidate data is implemented.

[0100] Another optional implementation is that the candidate data is automatically data-labeled, and in this implementation, the data processing module performs structured processing on the candidate data to obtain structured data, and data labels of the plurality of candidate data are obtained based on a preset labeling rule.

[0101] After data labeling is performed on each candidate data, based on the labeling information corresponding to each candidate data, the abnormal data in the plurality of candidate data is deleted to obtain training data, and the training data is stored to model meta information.

[0102] In the embodiment, data is imported by the data processing module, and the data is labeled to obtain training data, and then the training data is stored in the model meta information. Since the model meta information can be transmitted across platforms among the model development module, the model training module, the model deployment module and the model inference module, in the model training process, the training data generated by the data processing module does not need to be copied to the model training module, thereby improving the efficiency of model delivery.

[0103] Optionally, before the model meta information is obtained, the method comprises:

[0104] obtaining model development code;

[0105] compiling the model development code to generate a machine learning model;

[0106] obtaining model environment code;

[0107] debugging a training environment corresponding to the machine learning model based on the model environment code;

[0108] storing the model development code and the model environment code to model meta information;

[0109] The model development code is used for compiling the machine learning model, and the model environment code is used for compiling a model environment corresponding to the machine learning model.

[0110] As described above, the model delivery platform comprises a model development module and a model deployment module.

[0111] In the embodiment, the model development module obtains model development code input by an algorithm engineer, and compiles the model development code to generate a machine learning model. The model deployment module obtains model environment code input by the algorithm engineer, and debugs a training environment corresponding to the machine learning model based on the model environment code. Further, the model development code and the model environment code are stored to model meta information.

[0112] In the embodiment, the model development code and the model environment code are stored to model meta information. Since the model meta information can be transmitted across platforms among the model development module, the model training module, the model deployment module and the model inference module, in the model deployment process, there is no need to repeatedly build a model environment and repeatedly compile a machine learning model, thereby improving the efficiency of model delivery.

[0113] For the convenience of understanding the technical solutions set forth in all the above embodiments, please refer toFigure 3 , Figure 3 The "data preparation" part shown can be understood as a data processing module in the embodiments of the present application, the "model development" part can be understood as a model development module in the embodiments of the present application, the "model training" part can be understood as a model training module in the embodiments of the present application, the "model deployment" part can be understood as a model deployment module in the embodiments of the present application, and the "inference" part can be understood as a model inference module in the embodiments of the present application.

[0114] As shown in FIG. 1, the data processing module responds to an instruction sent by a data processor, the model development module, the model training module, the model deployment module and the model inference module respond to an instruction sent by an algorithm engineer, and the model meta information can be transmitted between the model processing module, the model development module, the model training module, the model deployment module and the model inference module. Figure 3

[0115] In the application scenario shown, the model training and the model deployment are both completed by the algorithm engineer, and the model training and the model deployment can be implemented without cross-platform, thereby improving the efficiency of model delivery. Figure 3

[0116] Optionally, the deploying the target machine learning model based on the model environment comprises:

[0117] receiving a computing power adjustment instruction and a capacity adjustment instruction;

[0118] redistributing, in response to the computing power adjustment instruction, a computing power resource corresponding to the target machine learning model in a target cluster; the target cluster is a cluster in which the target machine learning model is deployed;

[0119] redistributing, in response to the capacity adjustment instruction, a storage resource corresponding to the target machine learning model in the target cluster;

[0120] deploying the target machine learning model with the redistributed computing power resource and storage resource in the model environment.

[0121] In the process of model deployment, if a computing power adjustment instruction is received, then the computing power resource corresponding to the target machine learning model is redistributed based on the computing power adjustment instruction. Specifically, a mapping relationship between the target machine learning model and a computing node in the target cluster can be established, and the target machine learning model is deployed through the computing node with the mapping relationship.

[0122] In the process of model deployment, if a capacity adjustment instruction is received, then the storage resource corresponding to the target machine learning model is redistributed based on the capacity adjustment instruction. Specifically, the target machine learning model can be expanded or contracted.

[0123] ​​In the embodiment, during the model deployment process, the computing power resources and the storage resources of the target machine learning model are adjusted through the computing power adjustment instruction and the capacity adjustment instruction, one-key deployment of the model is realized, and the convenience of model deployment is improved.

[0124] Optionally, the method further includes:

[0125] receiving an account login instruction, the account login instruction carrying an account name and an account password;

[0126] In response to the account login instruction, the account name and the account password are verified.

[0127] In the case where the verification is passed and a read instruction is received, model meta information associated with the account name is read.

[0128] The model meta information includes training data, model development code, and model environment code.

[0129] In the embodiment, the model delivery platform can also set up a personal account, the personal account is associated with the model meta information, and different accounts are associated with different model meta information.

[0130] Optionally, the method is to receive an account login instruction, parse the account login instruction to obtain an account name and an account password, and verify the account name and the account password.

[0131] An optional verification method is to query the account name in a preset mapping table to obtain a password, and if the password is consistent with the account password, it is determined that the verification is passed. The mapping table stores the mapping relationship between the account name and the password.

[0132] In the case where the verification is passed, if a read quality indicating that the model meta information is obtained is received, the model meta information associated with the account name is obtained.

[0133] In the embodiment, the account is verified to obtain the model meta information corresponding to the account, so as to realize customization of the personal working directory of each account and enrich the personalized model delivery process.

[0134] Please refer to Figure 4 The model delivery platform 400 provided by the embodiment of the application includes a model training module 410, a model deployment module 420, and a model inference module 430. The model training module 410 is in communication connection with the model deployment module 420, and the model deployment module 420 is in communication connection with the model inference module 430.

[0135] The model training module 410 is configured to obtain model meta information.

[0136] parsing the model meta-information to obtain training data;

[0137] training a machine learning model based on the training data to obtain a trained target machine learning model;

[0138] a model deployment module 420, configured to parse the model meta-information to obtain a model environment corresponding to the machine learning model;

[0139] deploying the target machine learning model based on the model environment;

[0140] a model inference module 430, configured to enable the target machine learning model in a case where the deployment of the target machine learning model is completed.

[0141] In the embodiments of the present application, the model training module 410, the model deployment module 420, and the model inference module 430 can be operated by an algorithm engineer, that is, the algorithm engineer only needs to implement the processes of model training, model deployment, and model inference in one platform, so that the model training and model deployment can be implemented without cross-platform, and the efficiency of model delivery is improved.

[0142] Referring to Figure 5 The model delivery platform 400 provided in the embodiments of the present application further includes:

[0143] a data processing module 440, configured to import a data set; the data set includes a plurality of candidate data;

[0144] performing data labeling on each candidate data;

[0145] deleting abnormal data in the plurality of candidate data to obtain training data; the data labeling corresponding to the abnormal data represents that the candidate data is abnormal;

[0146] storing the training data to model meta-information.

[0147] Referring to Figure 5 The model delivery platform 400 provided in the embodiments of the present application further includes:

[0148] a model development module 450, configured to obtain model development code;

[0149] compiling the model development code to generate a machine learning model;

[0150] obtaining model environment code;

[0151] debugging a training environment corresponding to the machine learning model based on the model environment code;

[0152] store the model development code and the model environment code to model meta-information;

[0153] The model development code is used for compiling the machine learning model, and the model environment code is used for compiling a model environment corresponding to the machine learning model.

[0154] Optionally, the model deployment module 420 is specifically configured to:

[0155] receive a computing power adjustment instruction and a capacity adjustment instruction;

[0156] in response to the computing power adjustment instruction, re-allocate computing power resources corresponding to the target machine learning model in a target cluster; the target cluster is a cluster in which the target machine learning model is deployed;

[0157] in response to the capacity adjustment instruction, re-allocate storage resources corresponding to the target machine learning model in the target cluster;

[0158] deploy the target machine learning model re-allocated with computing power resources and storage resources in the model environment.

[0159] Optionally, the model delivery platform 400 is further configured to:

[0160] receive an account login instruction; the account login instruction carries an account name and an account password;

[0161] in response to the account login instruction, verify the account name and the account password;

[0162] in a case where the verification is passed and a reading instruction is received, read model meta-information associated with the account name;

[0163] The model meta-information includes training data, model development code, and model environment code.

[0164] To solve the above technical problems, the embodiments of the present application further provide a computer device. For details, please refer to Figure 6 , Figure 6 The basic structure block diagram of the computer device of the present embodiment is shown in FIG. 1.

[0165] The computer device 6 comprises a memory 61, a processor 62, and a network interface 63 which are communicatively connected by a system bus. It should be noted that only the computer device 6 with the memory 61, the processor 62, and the network interface 63 is shown in the figure, but it should be understood that all the components shown are not required to be implemented, and more or fewer components can be alternatively implemented. Among them, those skilled in the art can understand that the computer device herein is a device capable of automatically performing numerical calculation and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0166] The computer device can be a desktop computer, a notebook computer, a palm computer, a cloud server, and the like. The computer device can interact with a user through a keyboard, a mouse, a remote controller, a touchpad, a voice control device, and the like.

[0167] The memory 61 comprises at least one type of readable storage medium, including a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, and the like. In some embodiments, the memory 61 can be an internal storage unit of the computer device 6, such as a hard disk or a memory of the computer device 6. In other embodiments, the memory 61 can also be an external storage device of the computer device 6, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, and the like. Of course, the memory 61 can also include both the internal storage unit and the external storage device of the computer device 6. In the present embodiment, the memory 61 is generally used to store an operating system and various application software installed in the computer device 6, such as program codes of the model delivery method, etc. In addition, the memory 61 can also be used to temporarily store various data that have been output or will be output.

[0168] The processor 62 may, in some embodiments, be a Central Processing Unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 62 is generally used to control the overall operation of the computer device 6. In the present embodiment, the processor 62 is configured to execute program code stored in the memory 61 or to process data, such as to execute program code of the model delivery method.

[0169] The network interface 63 may, in some embodiments, be a wireless network interface or a wired network interface. The network interface 63 is generally used to establish a communication connection between the computer device 6 and other electronic devices.

[0170] The present application also provides another embodiment, i.e., to provide a computer readable storage medium storing the model delivery program, which can be executed by at least one processor to cause the at least one processor to perform the steps of the model delivery method as described above.

[0171] From the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment method can be realized by means of software and necessary general hardware online platform, of course, it can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes a plurality of instructions for causing a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present application.

[0172] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment, in which tasks are performed by remote processing devices connected by a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0173] Obviously, the above-described embodiments are only some embodiments but not all the embodiments of the present application, the preferred embodiments of the present application are shown in the drawings, but do not limit the patent scope of the present application. The present application can be implemented in many different forms, and conversely, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent replacements to some technical features therein. Any equivalent structure made by using the content of the specification and drawings, directly or indirectly applied to other related technical fields, is also within the patent protection scope of the present application.

Claims

1. A model delivery method characterized by comprising: The method comprises the following steps: acquiring model meta information; parsing the model meta information to obtain training data; training a machine learning model based on the training data to obtain a trained target machine learning model; parsing the model meta information to obtain a model environment corresponding to the machine learning model; deploying the target machine learning model based on the model environment; in the case that the deployment of the target machine learning model is completed, enabling the target machine learning model to perform inference on received service information; wherein the deployment of the target machine learning model based on the model environment comprises: receiving a computing power adjustment instruction and a capacity adjustment instruction; in response to the computing power adjustment instruction, reallocating computing power resources corresponding to the target machine learning model in a target cluster; the target cluster is a cluster where the target machine learning model is deployed; in response to the capacity adjustment instruction, reallocating storage resources corresponding to the target machine learning model in the target cluster; deploying the target machine learning model with the reallocated computing power resources and storage resources in the model environment; wherein, before the model meta information is acquired, the method comprises: acquiring model development code; compiling the model development code to generate a machine learning model; acquiring model environment code; debugging a training environment corresponding to the machine learning model based on the model environment code; storing the model development code and the model environment code to the model meta information; wherein the model development code is used to compile the machine learning model, and the model environment code is used to compile the model environment corresponding to the machine learning model.

2. The method of claim 1, wherein, Before the model meta information is acquired, the method comprises: importing a data set, the data set comprising a plurality of candidate data; performing data labeling on each candidate data; deleting abnormal data in the plurality of candidate data to obtain training data; the data labeling corresponding to the abnormal data represents that the candidate data is abnormal; storing the training data to the model meta information.

3. The method according to any one of claims 1-2, characterized in that, The method further comprises: receiving an account login instruction; the account login instruction carries an account name and an account password; in response to the account login instruction, verifying the account name and the account password; in the case that the verification is passed and a reading instruction is received, reading model meta information associated with the account name; wherein the model meta information comprises training data, model development code and model environment code.

4. A model delivery platform characterized by, The method comprises a model training module, a model deployment module and a model inference module, the model training module and the model deployment module are in communication connection, and the model deployment module and the model inference module are in communication connection: the model training module is used for acquiring model meta information; parsing the model meta information to obtain training data; training a machine learning model based on the training data to obtain a trained target machine learning model; the model deployment module is used for parsing the model meta information to obtain a model environment corresponding to the machine learning model; deploying the target machine learning model based on the model environment; The model inference module is configured to, in a case where the target machine learning model is deployed, enable the target machine learning model to perform inference on received service information. The model deployment module is configured to: receive a computing power adjustment instruction and a capacity adjustment instruction; in response to the computing power adjustment instruction, re-allocate computing power resources corresponding to the target machine learning model in a target cluster; the target cluster is a cluster in which the target machine learning model is deployed; in response to the capacity adjustment instruction, re-allocate storage resources corresponding to the target machine learning model in the target cluster; deploy the target machine learning model with the re-allocated computing power resources and storage resources in the model environment. The model delivery platform further comprises: a model development module configured to obtain model development code; compile the model development code to generate a machine learning model; obtain model environment code; debug a training environment corresponding to the machine learning model based on the model environment code; store the model development code and the model environment code to model meta information; The model development code is used to compile the machine learning model, and the model environment code is used to compile a model environment corresponding to the machine learning model.

5. The model delivery platform of claim 4, wherein, The model delivery platform further comprises: a data processing module configured to import a data set; the data set comprises a plurality of candidate data; perform data labeling on each candidate data; delete abnormal data in the plurality of candidate data to obtain training data; the data labeling of the abnormal data represents that the candidate data is abnormal; store the training data to model meta information.

6. A computer device, comprising: A computer readable storage medium stores a computer program, and a processor executes the computer program to implement the steps of the model delivery method according to any one of claims 1 to 3.

7. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and a processor executes the computer program to implement the steps of the model delivery method according to any one of claims 1 to 3.

Images