CN114117915A - Model deployment method and device and electronic equipment - Google Patents
Model deployment method and device and electronic equipment Download PDFInfo
- Publication number
- CN114117915A CN114117915A CN202111423590.5A CN202111423590A CN114117915A CN 114117915 A CN114117915 A CN 114117915A CN 202111423590 A CN202111423590 A CN 202111423590A CN 114117915 A CN114117915 A CN 114117915A
- Authority
- CN
- China
- Prior art keywords
- model
- server
- information
- corpus
- file
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Debugging And Monitoring (AREA)
Abstract
The application provides a model deployment method, a model deployment device and electronic equipment, wherein the method comprises the following steps: receiving request information sent by a user side; the request information carries information of a model and information of a corpus; generating corpus data according to the information of the corpus; storing the corpus data to a first server, wherein the corpus data is used for model training; sending the information of the model to a second server, wherein the information of the model is used for determining the model to be trained; under the condition that the second server is determined to finish training the model, reading a model file of the model from the first server, wherein the model file is stored to the first server by the second server; and carrying out model deployment according to the model file. The method reduces the cost of model iteration, improves the efficiency of model deployment, and can realize the rapid iteration, deployment and application of the model.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a model deployment method and apparatus, and an electronic device.
Background
In the field of artificial intelligence, models are often used to assist in the implementation of various functions in a scene, in various application scenarios. For example, in the field of intelligent customer service, a model is generally used to assist in implementing functions such as text classification, entity extraction, QA question answering, and in the field of machine translation, a model is generally used to assist in implementing functions such as text analysis and translation.
In some application scenarios, the service content is updated faster, and the applied model is required to be iterated rapidly according to the change of the service content. The traditional model deployment method has the disadvantages of complicated steps, low model deployment efficiency, high model iteration cost and the possibility of model training and deployment failure.
Disclosure of Invention
Therefore, the application provides a model deployment method, a model deployment device and electronic equipment, so as to at least solve the technical problems of high model iteration cost and low model deployment efficiency in the prior art.
An embodiment of a first aspect of the present application provides a model deployment method, including:
receiving request information sent by a user side; the request information carries information of a model and information of a corpus;
generating corpus data according to the information of the corpus;
storing the corpus data to a first server, wherein the corpus data is used for model training;
sending the information of the model to a second server, wherein the information of the model is used for determining the model to be trained;
under the condition that the model is completely trained by the second server, reading a model file of the model from the first server, wherein the model file is stored to the first server by the second server;
and carrying out model deployment according to the model file.
Optionally, the method further comprises:
reading an index file of the model from the first server, wherein the index file is obtained by testing the model by the second server and is stored to the first server;
and analyzing the index file to obtain index data in the index file, so that the user side can access the index data through a set interface.
Optionally, the generating corpus data according to the information of the corpus includes:
under the condition that the language material information contains the test language material appointed by the user side, generating a test sample in the language material data according to the test language material;
and generating a training sample in the corpus data according to the residual corpus information in the corpus information.
Optionally, the generating corpus data according to the information of the corpus includes:
under the condition that the corpus information does not have the test corpus specified by the user, dividing the corpus information into a first part and a second part according to a preset rule;
generating a training sample in the corpus data according to the first part of the corpus information;
and generating a test sample in the corpus data according to the second part of the corpus information.
Optionally, the generating corpus data according to the information of the corpus further includes: and preprocessing the corpus information.
Optionally, after the sending the information of the model to the second server, the method further includes:
periodically sending a status query request to the second server;
and determining whether the state of the second server is that the model is trained completely according to the state information returned by the second server.
Optionally, the information of the model includes a training algorithm identifier of the model and a model identifier of the model;
the training algorithm identifies a training algorithm for instructing the second server to determine the model.
An embodiment of a second aspect of the present application provides a model deployment apparatus, including:
the receiving unit is used for receiving request information sent by a user side; the request information carries information of a model and information of a corpus;
the processing unit is used for generating corpus data according to the information of the corpus;
the storage unit is used for storing the corpus data to a first server, wherein the corpus data is used for model training;
the sending unit is used for sending the information of the model to a second server, wherein the information of the model is used for determining the model to be trained;
a reading unit, configured to read a model file of the model from the first server when it is determined that the model is trained by the second server, where the model file is stored in the first server by the second server;
and the deployment unit is used for carrying out model deployment according to the model file.
Optionally, the reading unit is further configured to:
reading an index file of the model from the first server, wherein the index file is obtained by testing the model by the second server and is stored to the first server;
and analyzing the index file to obtain index data in the index file, so that the user side can access the index data through a set interface.
Optionally, the processing unit is specifically configured to:
under the condition that the language material information contains the test language material appointed by the user side, generating a test sample in the language material data according to the test language material;
and generating a training sample in the corpus data according to the residual corpus information in the corpus information.
Optionally, the processing unit is further specifically configured to:
under the condition that the corpus information does not have the test corpus specified by the user, dividing the corpus information into a first part and a second part according to a preset rule;
generating a training sample in the corpus data according to the first part of the corpus information;
and generating a test sample in the corpus data according to the second part of the corpus information.
Optionally, the processing unit is further specifically configured to: and preprocessing the corpus information.
Optionally, the sending unit is further specifically configured to:
periodically sending a status query request to the second server;
and determining whether the state of the second server is that the model is trained completely according to the state information returned by the second server.
Optionally, the information of the model includes a training algorithm identifier of the model and a model identifier of the model;
the training algorithm identifies a training algorithm for instructing the second server to determine the model.
An embodiment of a third aspect of the present application provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of the embodiment of the first aspect of the present application.
A fourth aspect of the present application provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method of the first aspect of the present application.
According to the model deployment method provided by the embodiment of the application, request information sent by a user side is received; the request information carries information of a model and information of a corpus; generating corpus data according to the information of the corpus; storing the corpus data to a first server, wherein the corpus data is used for model training; sending the information of the model to a second server, wherein the information of the model is used for determining the model to be trained; reading a model file of the model from the first server when the state of the second server is determined to be that the model is trained completely, wherein the model file is stored to the first server by the second server; and carrying out model deployment according to the model file. The method reduces the cost of model iteration, improves the efficiency of model deployment, and can realize the rapid iteration, deployment and application of the model.
Additional aspects and advantages of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a schematic flowchart of a model deployment method according to an embodiment of the present disclosure;
FIG. 2 is a block diagram of a system architecture according to an embodiment of the present disclosure;
fig. 3 is a schematic flowchart of another model deployment method provided in an embodiment of the present application;
fig. 4 is a schematic structural diagram of a model deployment apparatus provided in an embodiment of the present application;
fig. 5 is a schematic structural diagram of an embodiment of an electronic device according to the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining the present application and should not be construed as limiting the present application.
The following describes a model deployment method and apparatus according to an embodiment of the present application with reference to the drawings.
Fig. 1 is a schematic flowchart of a model deployment method provided in an embodiment of the present application, where the method provided in this embodiment may be executed by various electronic devices with data processing capability, and the electronic device executing the method of this embodiment is not limited herein, as shown in fig. 1, the method includes the following steps:
In some embodiments, the user terminal generates and transmits the request message through the operation of the user.
Optionally, the user performs an interactive operation through an interface of the user side, and generates the request information through an input window and an interactive button provided by the interface. As an exemplary implementation manner, according to a requirement of a user on a model, information of a corpus and information of the model are input, and a "one-key training" on an interface is clicked to generate and send request information, where the request information carries the information of the model and the information of the corpus.
The information of the linguistic data refers to information of the linguistic data which is input by a user and used for training and/or testing the model, and the user can input the linguistic data singly or in batches, can input files for storing the linguistic data, and can input addresses of the files for storing the linguistic data.
The information of the model refers to basic information that can determine the model to be trained, and may include a model identifier and a training algorithm identifier of the model. According to the information of the model, the model to be trained can be determined, and the training algorithm to be selected for training the model can be determined.
In some embodiments, after receiving the request information sent by the user side, a model record corresponding to the request information may be created, and the record may include a model identifier, a record of a model version, a training record, and a corpus record.
And 102, generating corpus data according to the information of the corpus.
Optionally, the corpus data is generated according to the information of the corpus carried in the received request information.
In some embodiments, the corpus data may be generated based on a user-entered corpus.
In some embodiments, a file storing the corpus input by the user may be read, and corpus data may be generated according to the corpus in the file.
In some embodiments, the file storing the corpus input by the user may be read according to the address of the file storing the corpus input by the user, and the corpus data may be generated according to the corpus in the file.
Optionally, the corpus data includes training samples and testing samples, where the training samples are used to train the model, and the testing samples are used to test the trained model.
Optionally, the user may specify the test corpus in the input corpus information, or may not specify, which is not limited in this embodiment.
And 103, storing the corpus data to a first server, wherein the corpus data is used for model training.
The first server may be a physical server, a cloud server, or a server of a distributed system. Alternatively, the first server may be an OSS (Object Storage Service) platform.
In some embodiments, the generated corpus data may be written to a file and stored to the first server. Optionally, the training samples and the testing samples are written into a file and stored to the first server, respectively.
In some embodiments, the second server reads the corpus data from the first server for model training using the corpus data.
The second server may be a physical server, a cloud server, or a server of a distributed system. The second server can read the corpus data from the first server, and train the model by using the read corpus data.
In some embodiments, the second server reads a file in the first server, in which the corpus data is stored, and trains the model using the corpus data in the read file. Optionally, the second server reads the file in which the training sample is stored and the file in which the test sample is stored in the first server, trains the model by using the training sample in the read file, and tests the model by using the test sample in the read file.
And 104, sending the information of the model to a second server, wherein the information of the model is used for determining the model to be trained.
And the second server determines the model to be trained according to the information of the model.
In some embodiments, the second server has a plurality of training algorithms stored therein. Optionally, the second server determines a model to be trained and a corresponding training algorithm according to the information of the model, and trains the model.
And 105, under the condition that the model is completely trained by the second server, reading a model file of the model from the first server, wherein the model file is stored to the first server by the second server.
And the second server determines a model to be trained according to the information of the model and trains the model by utilizing the corpus data. After the model training is completed, the second server generates a model file of the trained model and stores the model file to the first server.
And reading the model file from the first server under the condition that the second server is determined to finish training the model.
In some embodiments, it is determined whether the model is trained based on the state information of the second server. Optionally, the state information of the second server includes that the model training is not completed and the model training is completed.
In some embodiments, a query request is sent to the second server, and whether the model is trained is determined according to a response returned by the second server.
And 106, deploying the model according to the model file.
And carrying out model deployment according to the model file read in the first server.
Optionally, the model file is loaded to provide a service interface so that the model can be applied to complete the deployment of the model.
In order to better understand the method for deploying the model disclosed in the embodiment of the present application, a system to which the embodiment of the present application is applicable is described with reference to fig. 2.
Referring to fig. 2, fig. 2 is a schematic diagram of a system architecture according to an embodiment of the present disclosure. The system may include a user terminal 201, a first server 202, a second server 203, and an electronic device 204. The method disclosed by the embodiment of the application is executed by the electronic device 204. The number and the form of the devices shown in fig. 2 are only examples and do not limit the embodiments of the present disclosure, and the practical application may include two or more clients, two or more electronic devices, and two or more servers. The system shown in fig. 2 includes a user terminal 201, a first server 202, a second server 203 and an electronic device 204.
In the system shown in the figure, a user terminal 201 sends request information to an electronic device 204, where the request information carries information of a model and information of a corpus, so that the electronic device 204 generates corpus data according to the information of the corpus, and stores the corpus data in a first server 202, so that a second server 203 reads the corpus data from the first server 202, and the electronic device 204 also sends the information of the model to the second server 203, so that the second server 203 trains the model based on the read corpus data. After the second server 203 finishes training the model, the second server 203 stores the model file in the first server 202, so that the electronic device 204 acquires the model file from the first server 202 and deploys the model according to the model file.
According to the model deployment method provided by the embodiment of the application, request information sent by a user side is received; the request information carries information of a model and information of a corpus; generating corpus data according to the information of the corpus; storing the corpus data to a first server, wherein the corpus data is used for a second server to read the corpus data from the first server for model training; sending the information of the model to the second server, wherein the information of the model is used for determining the model to be trained; reading a model file of the model from the first server when the state of the second server is determined to be that the model is trained completely, wherein the model file is stored to the first server by the second server; and carrying out model deployment according to the model file. According to the method, the steps of model training deployment are simplified, the cost of model iteration is reduced, the efficiency of model deployment is improved, and the rapid iteration, deployment and application of the model can be realized through the method.
Fig. 3 is a schematic flowchart of another model deployment method provided in an embodiment of the present application, and as shown in fig. 3, the model deployment method includes the following steps:
Wherein the pre-processing may include at least one of digital normalization, case and case conversion, full half angle conversion. And the corpus information is preprocessed, so that the trained model has a better effect.
In some embodiments, if the corpus information includes a test corpus specified by a user, a test sample in the corpus data is generated according to the test corpus specified by the user, and a training sample in the corpus data is generated according to the remaining corpus information in the corpus information.
Alternatively, the user may designate a part of the corpus as the test corpus in the corpus information.
Optionally, the user may specify a certain proportion of the corpus in the corpus information as the test corpus, where the proportion may be a proportion input by the user or a proportion selected by the user according to an option.
In some embodiments, the corpus information does not include a test corpus specified by a user, the corpus information is divided into a first part and a second part according to a preset rule, a training sample is generated according to the corpus information of the first part, and a test sample is generated according to the corpus information of the second part.
Optionally, the corpus information may be divided according to a preset rule, the corpus may be divided according to a default ratio, the ratio may be 7:3, that is, 70% of the corpus information is used as a first part to generate a training sample, and 30% of the corpus information is used as a second part to generate a test sample. The proportion can be adjusted according to different scenes, requirements and data volume of the corpus information. The proportion of the training samples in the corpus information is 2/3-4/5.
Optionally, the corpus information may be divided into a third part as the verification data.
Optionally, the training samples and the testing samples in the corpus data are stored to the first server. And the second server reads the training samples and the test samples in the first server, and trains and tests the model.
In some embodiments, a training request is sent to the second server, where the training request carries an address where corpus data is stored in the first server. And the second server reads the corpus data stored in the first server according to the address of the corpus data in the training request, and performs model training by using the corpus data.
And step 304, sending information of the model to the second server, wherein the information of the model is used for determining the model to be trained.
In some embodiments, the information of the model includes a training algorithm identification of the model and a model identification of the model. Wherein the training algorithm of the model identifies a training algorithm used to instruct the second server to determine the model.
Optionally, the second server stores a plurality of training algorithms, and the training algorithms can be trained for different models. And the second server determines a corresponding training algorithm according to the training algorithm identification in the information of the model, and trains the model.
In some embodiments, the information of the model may also be carried by the training request in step 303 and sent to the second server.
Periodically, a status query request is sent to the second server, step 305.
Optionally, the status query request is used to determine the status of the second server.
And step 306, determining whether the state of the second server is that the model training is completed according to the state information returned by the second server.
If yes, go to step 307, otherwise, go back to step 305 to continue polling.
Optionally, the status information includes: model training is complete and model training is incomplete.
And 307, under the condition that the state of the second server is determined to be that the model training is completed, reading a model file of the model from the first server, wherein the model file is stored to the first server by the second server.
Optionally, after receiving the state information returned by the second server as that the model training is completed, modifying the training state in the training record to be completed.
The index data of the model is an index that can be used to evaluate the effect of the model, and generally includes accuracy, precision, recall, confusion matrix, mutual information, and so on.
In the embodiment of the application, different index data can be determined according to different types of models, the types of the indexes can be determined in a training algorithm, and the indexes can include at least one of accuracy, precision, recall, confusion matrix and mutual information.
And the second server tests the trained model by using the test sample in the corpus data to obtain index data of the model, writes the index data into the index file, and stores the index data in the first server.
And reading the index file from the first server under the condition that the state of the second server is determined to be that the model is trained completely.
Optionally, the index file is parsed to obtain index data, the index data is stored in the database, and an interface is provided, so that the user side can access the index data of the model through the interface.
And 310, deploying the model according to the model file.
The method provided by the embodiment of the application comprises the steps of receiving request information sent by a user side, wherein the request information carries information of a model and information of a corpus, preprocessing the corpus information, generating corpus data according to the preprocessed corpus information, storing the corpus data in a first server, wherein the corpus data is used for model training, sending the information of the model to a second server, wherein the information of the model is used for determining the model to be trained, periodically sending a state query request to the second server, determining whether the state of the second server is that the model training is completed according to state information returned by the second server, reading a model file of the model from the first server under the condition that the state of the second server is that the model training is completed, wherein the model file is stored to the first server by the second server, and reading an index file of the model from the first server, wherein the index file is stored to the first server by the second server, analyzing the index file, and obtaining index data in the index file, so that a user end can access the index data through a set interface, and model deployment is carried out according to the model file. The method can complete the training, deployment and application of the model after receiving the request information sent by the user side, simplifies the steps of model training and deployment, reduces the cost of model iteration, improves the efficiency of model deployment, and realizes the rapid iteration, deployment and application of the model.
The application also provides a model deployment device. Fig. 4 is a schematic structural diagram of a model deployment device according to an embodiment of the present application.
As shown in fig. 4, the deployment apparatus of the model includes: receiving unit 410, processing unit 420, storage unit 430, transmitting unit 440, reading unit 450, and deployment unit 460.
A receiving unit 410, configured to receive request information sent by a user end; the request information carries information of a model and information of a corpus;
a processing unit 420, configured to generate corpus data according to the information of the corpus;
a storage unit 430, configured to store the corpus data to a first server, where the corpus data is used for model training;
a sending unit 440, configured to send information of the model to a second server, where the information of the model is used to determine a model to be trained;
a reading unit 450, configured to read a model file of the model from the first server when it is determined that the model is trained by the second server, where the model file is stored to the first server by the second server;
and the deployment unit 460 is configured to perform model deployment according to the model file.
In some embodiments, the reading unit 450 is further configured to: reading a metric file of the model from the first server, wherein the metric file is stored by the second server to the first server; and analyzing the index file to obtain index data in the index file, so that the user side can access the index data through a set interface.
In some embodiments, the processing unit 420 is specifically configured to: under the condition that the language material information contains the test language material appointed by the user side, generating a test sample in the language material data according to the test language material; and generating a training sample in the corpus data according to the residual corpus information in the corpus information.
In some embodiments, the processing unit 420 is further specifically configured to: under the condition that the corpus information does not have the test corpus specified by the user, dividing the corpus information into a first part and a second part according to a preset rule; generating a training sample in the corpus data according to the first part of the corpus information; and generating a test sample in the corpus data according to the second part of the corpus information.
In some embodiments, the processing unit 420 is further specifically configured to: and preprocessing the corpus information.
In some embodiments, the sending unit 440 is further specifically configured to: periodically sending a status query request to the second server; and determining whether the state of the second server is that the model is trained completely according to the state information returned by the second server.
In some embodiments, the information of the model includes a training algorithm identification of the model and a model identification of the model; the training algorithm identifies a training algorithm for instructing the second server to determine the model.
It should be noted that the foregoing explanation of the method embodiment is also applicable to the apparatus of this embodiment, and is not repeated herein.
The device provided by the embodiment of the application, by receiving request information sent by a user side, wherein the request information carries information of a model and information of a corpus, preprocessing the corpus information, generating corpus data according to the preprocessed corpus information, storing the corpus data in a first server, wherein the corpus data is used for model training, sending the information of the model to a second server, wherein the information of the model is used for determining the model to be trained, periodically sending a state query request to the second server, determining whether the state of the second server is that the model training is completed according to state information returned by the second server, and reading a model file of the model from the first server when the state of the second server is that the model training is completed, wherein the model file is stored to the first server by the second server, and reading an index file of the model from the first server, wherein the index file is stored to the first server by the second server, analyzing the index file, and obtaining index data in the index file, so that a user end can access the index data through a set interface, and model deployment is carried out according to the model file. Therefore, after the request information sent by the user side is received, the training, deployment and application of the model can be completed, the steps of model training and deployment are simplified, the cost of model iteration is reduced, the efficiency of model deployment is improved, and the rapid iteration, deployment and application of the model are realized.
An embodiment of the present application further provides an electronic device, which includes the apparatus according to any of the foregoing embodiments.
Fig. 5 is a schematic structural diagram of an embodiment of an electronic device provided in the present application, which may implement the flows of the embodiments shown in fig. 1 and 3 of the present invention, and as shown in fig. 5, the electronic device may include:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any of the intelligent response methods described above.
The embodiment of the application also provides a non-transitory computer readable storage medium storing computer instructions, wherein the computer instructions are used for causing a computer to execute any one of the intelligent answering methods.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A method for deploying a model, comprising:
receiving request information sent by a user side; the request information carries information of a model and information of a corpus;
generating corpus data according to the information of the corpus;
storing the corpus data to a first server, wherein the corpus data is used for model training;
sending the information of the model to a second server, wherein the information of the model is used for determining the model to be trained;
under the condition that the second server finishes training the model, reading a model file of the model from the first server, wherein the model file is stored to the first server by the second server;
and carrying out model deployment according to the model file.
2. The method of claim 1, further comprising:
reading an index file of the model from the first server, wherein the index file is obtained by testing the model by the second server and is stored to the first server;
and analyzing the index file to obtain index data in the index file, so that the user side can access the index data through a set interface.
3. The method according to claim 1, wherein said generating corpus data according to said corpus information comprises:
under the condition that the language material information contains the test language material appointed by the user side, generating a test sample in the language material data according to the test language material;
and generating a training sample in the corpus data according to the residual corpus information in the corpus information.
4. The method according to claim 1, wherein said generating corpus data according to said corpus information comprises:
under the condition that the corpus information does not have the test corpus specified by the user, dividing the corpus information into a first part and a second part according to a preset rule;
generating a training sample in the corpus data according to the first part of the corpus information;
and generating a test sample in the corpus data according to the second part of the corpus information.
5. The method according to claim 3 or 4, wherein said generating corpus data according to said corpus information further comprises:
and preprocessing the corpus information.
6. The method according to any one of claims 1-4, wherein after sending the information of the model to the second server, further comprising:
periodically sending a status query request to the second server;
and determining whether the state of the second server is that the model is trained completely according to the state information returned by the second server.
7. The method according to any one of claims 1-4, wherein the information of the model comprises a training algorithm identification of the model and a model identification of the model;
the training algorithm identifies a training algorithm for instructing the second server to determine the model.
8. An apparatus for deploying a model, comprising:
the receiving unit is used for receiving request information sent by a user side; the request information carries information of a model and information of a corpus;
the processing unit is used for generating corpus data according to the information of the corpus;
the storage unit is used for storing the corpus data to a first server, wherein the corpus data is used for model training;
the sending unit is used for sending the information of the model to the second server, wherein the information of the model is used for determining the model to be trained;
a reading unit, configured to read a model file of the model from the first server when the second server finishes training the model, where the model file is stored to the first server by the second server;
and the deployment unit is used for carrying out model deployment according to the model file.
9. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.
10. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111423590.5A CN114117915A (en) | 2021-11-26 | 2021-11-26 | Model deployment method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111423590.5A CN114117915A (en) | 2021-11-26 | 2021-11-26 | Model deployment method and device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114117915A true CN114117915A (en) | 2022-03-01 |
Family
ID=80370231
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111423590.5A Pending CN114117915A (en) | 2021-11-26 | 2021-11-26 | Model deployment method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114117915A (en) |
-
2021
- 2021-11-26 CN CN202111423590.5A patent/CN114117915A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108073519B (en) | Test case generation method and device | |
| CN111262759B (en) | Internet of things platform testing method, device, equipment and storage medium | |
| CN109360550A (en) | Test method, device, equipment and the storage medium of voice interactive system | |
| CN110554958A (en) | Graph database testing method, system, device and storage medium | |
| CN112667778A (en) | Information input method, device, equipment and storage medium combining RPA and AI | |
| CN110727580A (en) | Response data generation method, full-flow interface data processing method and related equipment | |
| CN105068935A (en) | Method and device for processing software test result | |
| CN113138934B (en) | Automatic test method, medium, device and computing equipment | |
| CN111104542A (en) | Part identification management method and device | |
| CN111857103B (en) | Vehicle diagnosis method, device, equipment and storage medium | |
| CN113934758A (en) | Vehicle fault repairing method and device, vehicle-mounted terminal, server and storage medium | |
| CN114185766A (en) | Code detection method and device, electronic equipment and storage medium | |
| CN113448730A (en) | Service processing method and device, computer equipment and storage medium | |
| CN113535538B (en) | Method, device, electronic equipment and storage medium for automatically testing application full link | |
| CN111143650B (en) | Method, device, medium and electronic equipment for acquiring page data | |
| CN111124894B (en) | Code coverage rate processing method and device and computer equipment | |
| CN110008074B (en) | Method, device and equipment for automatically testing and inquiring upper-layer interface of hardware information | |
| CN114117915A (en) | Model deployment method and device and electronic equipment | |
| CN115269673A (en) | Business big data processing method and system applying artificial intelligence | |
| CN114141302A (en) | Test method of solid state disk and electronic equipment | |
| CN110730342A (en) | Video quality analysis method and device, server and terminal | |
| CN114338846A (en) | Message testing method and device | |
| CN110618881A (en) | Method and system for simulating message queue | |
| CN110750409A (en) | Method, device and system for realizing test | |
| CN114168468B (en) | Software testing system and control method thereof |
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 |