CN114020432A - Task exception handling method and device and task exception handling system - Google Patents
Task exception handling method and device and task exception handling system Download PDFInfo
- Publication number
- CN114020432A CN114020432A CN202111308542.1A CN202111308542A CN114020432A CN 114020432 A CN114020432 A CN 114020432A CN 202111308542 A CN202111308542 A CN 202111308542A CN 114020432 A CN114020432 A CN 114020432A
- Authority
- CN
- China
- Prior art keywords
- target
- task
- exception
- training
- test case
- 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
- 238000000034 method Methods 0.000 title claims abstract description 84
- 238000012549 training Methods 0.000 claims abstract description 282
- 238000012545 processing Methods 0.000 claims abstract description 192
- 230000002159 abnormal effect Effects 0.000 claims abstract description 139
- 238000012360 testing method Methods 0.000 claims description 175
- 238000012795 verification Methods 0.000 claims description 87
- 230000005856 abnormality Effects 0.000 claims description 59
- 230000008569 process Effects 0.000 claims description 38
- 238000004891 communication Methods 0.000 description 10
- 230000008439 repair process Effects 0.000 description 10
- 238000004590 computer program Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000033772 system development Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/485—Task life-cycle, e.g. stopping, restarting, resuming execution
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
Abstract
The embodiment of the invention provides a task exception handling method and device and electronic equipment, and is applied to the technical field of algorithm training. The method comprises the following steps: when the target training task in the algorithm training system is detected to be abnormal in operation, acquiring a target abnormal reason causing the abnormal operation of the target training task; judging whether a preset processing strategy corresponding to the target abnormal reason exists or not; if the target training task exists, processing the abnormity occurring in the operation of the target training task according to a processing strategy; and if the configuration operation does not exist, outputting a notification message aiming at the target training task, and storing a processing strategy which is configured by the configuration operation and corresponds to the target abnormal reason when the configuration operation aiming at the target abnormal reason is received. By the scheme, the efficiency of task exception handling can be improved.
Description
Technical Field
The invention relates to the technical field of algorithm training, in particular to a task exception handling method, a task exception handling device and a task exception handling system.
Background
The algorithm training system is a back-end system that generates models by training algorithms. The algorithm may be a machine learning algorithm such as CNN (Convolutional Neural Network).
For an algorithm training system, it is most important to ensure the success rate of a training task, wherein the training task refers to: and (3) processing the data with the specific format through an algorithm program, and training to obtain a model process. Therefore, it is very important to handle the exception when the training task runs abnormally. In the related art, a manual processing mode is mostly adopted, that is, after the training task runs abnormally, a worker needs to be informed to manually process the abnormal running of the training task so as to recover the normal running of the algorithm training system.
However, with the enlargement of the scale of the algorithm training system and the increase of the target training tasks, the frequency of the target training tasks having abnormalities in the algorithm training process is higher and higher, and the efficiency is lower only by a manual processing mode.
Disclosure of Invention
The embodiment of the invention aims to provide a task exception handling method, a task exception handling device and a task exception handling system, so as to improve exception handling efficiency. The specific technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides a method for processing a task exception, where the method includes:
when the abnormal operation of a target training task in an algorithm training system is detected, acquiring a target abnormal reason causing the abnormal operation of the target training task;
judging whether a preset processing strategy corresponding to the target abnormal reason exists or not;
if the target training task exists, processing the abnormity occurring in the operation of the target training task according to the processing strategy;
and if not, outputting a notification message aiming at the target training task, and storing a processing strategy which is configured by the configuration operation and corresponds to the target abnormal reason when the configuration operation aiming at the target abnormal reason is received.
Optionally, before the outputting the notification message for the target training task, the method further includes:
generating a test case which contains the target abnormal reason and has an incomplete case state based on the training data of the target training task;
the storing of the processing policy corresponding to the target exception cause configured by the configuration operation includes:
and in the generated test case, setting a processing strategy which is configured by the configuration operation and corresponds to the target abnormal reason, and setting the case state of the generated test case to be a completion state.
Optionally, the determining whether a preset processing policy corresponding to the target abnormal reason exists includes
Judging whether a first test case which contains the target abnormal reason and is in a finished state exists or not;
if so, judging that a preset processing strategy corresponding to the target abnormal reason exists;
and if not, judging that a preset processing strategy corresponding to the target abnormal reason does not exist.
Optionally, the processing the exception occurring during the operation of the target training task according to the processing strategy includes:
and processing the abnormity occurring in the operation of the target training task according to the processing strategy in the first test case.
Optionally, each test case further includes: verification data for verifying whether processing for an exception caused by the included exception cause is valid;
after the processing the exception occurring during the operation of the target training task according to the processing strategy in the first test case, the method further includes:
running a first verification task based on verification data in the first test case; wherein each verification task is used for verifying whether the processing aiming at the exception is effective or not;
and if the first verification task runs normally, determining that the abnormal processing of the target training task is completed.
Optionally, the method further includes:
if the first verification task runs abnormally, displaying the abnormality of the target training task during running, and prompting to process the abnormality of the target training task during running;
after the exception is processed, returning to execute the step of running a first verification task based on the verification data in the first test case until the first verification task runs normally;
after the first verification task runs normally, if an updating operation of the processing strategy for the target abnormal reason is received, updating the processing strategy in the first test case according to the updating operation.
Optionally, before setting, in the generated test case, a processing policy configured by the configuration operation and corresponding to the target exception cause, and setting a case state of the generated test case to a complete state, the method further includes:
running a second verification task based on verification data in the generated test case;
the setting, in the generated test case, a processing policy corresponding to the target abnormality cause and configured by the configuration operation, and the setting of the case state of the generated test case to a completion state include:
if the second verification task runs normally, setting a processing strategy configured by the configuration operation and corresponding to the target abnormal reason in the generated test case, and setting the case state of the generated test case to be a completion state.
Optionally, the method further includes:
if the second verification task runs abnormally, displaying the abnormality of the target training task, and prompting to process the abnormality of the target training task;
and after the exception processing of the second verification task is completed, returning to execute the verification data in the generated test case, and running the second verification task until the second verification task runs normally.
In a second aspect, an embodiment of the present invention provides a task exception handling apparatus, where the apparatus includes:
the abnormal reason acquisition module is used for acquiring a target abnormal reason causing the abnormal operation of the target training task when the abnormal operation of the target training task in the algorithm training system is detected;
the processing strategy judging module is used for judging whether a preset processing strategy corresponding to the target abnormal reason exists or not;
the first exception handling module is used for handling the exception occurring in the operation of the target training task according to the handling strategy if the exception occurs;
and the second exception handling module is used for outputting a notification message aiming at the target training task if the exception does not exist, and storing a handling strategy which is configured by the configuration operation and corresponds to the target exception reason when the configuration operation aiming at the target exception reason is received.
In a third aspect, an embodiment of the present invention provides a task exception handling system, where the task exception handling system includes: training a server side and an exception handling side; the training server is deployed with an algorithm training system;
the training server is used for operating a target training task in the algorithm training system;
the abnormity processing terminal is used for acquiring a target abnormity reason causing the abnormity of the operation of the target training task from the training server terminal when the abnormity of the operation of the target training task is detected; judging whether a preset processing strategy corresponding to the target abnormal reason exists or not; if the target training task exists, controlling the training server according to the processing strategy so as to process the abnormity occurring in the operation of the target training task; and if not, outputting a notification message aiming at the target training task, and storing a processing strategy which is configured by the configuration operation and corresponds to the target abnormal reason when the configuration operation aiming at the target abnormal reason is received.
The embodiment of the invention has the following beneficial effects:
in the task exception handling method provided by the embodiment of the invention, when it is detected that a target training task in an algorithm training system runs abnormally, a target exception reason causing the target training task to run abnormally can be acquired, and then whether a preset handling strategy corresponding to the target exception reason exists is judged, if yes, the exception occurring in the running of the target training task is handled according to the handling strategy, if not, a notification message aiming at the target training task is output, and when configuration operation aiming at the target exception reason is received, the handling strategy corresponding to the target exception reason and configured by the configuration operation is stored. When a processing strategy corresponding to the target exception reason exists, the exception can be processed according to the processing strategy, so that the exception is automatically processed, and the exception processing efficiency is improved.
Of course, not all of the advantages described above need to be achieved at the same time in the practice of any one product or method of the invention.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other embodiments can be obtained by using the drawings without creative efforts.
Fig. 1 is a schematic structural diagram of a task exception handling system according to an embodiment of the present invention;
FIG. 2 is a flowchart of a task exception handling method according to an embodiment of the present invention;
FIG. 3 is a flowchart of a task exception handling method according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a task exception handling apparatus according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to improve the efficiency of exception handling, embodiments of the present invention provide a task exception handling method, a task exception handling device, and a task exception handling system.
It should be noted that the embodiments of the present invention can be applied to various electronic devices, such as personal computers, servers, smart mobile devices, and other devices having data processing capability. Moreover, the task exception handling method provided by the embodiment of the invention can be realized in a software, hardware or a combination of software and hardware.
As shown in fig. 1, a schematic diagram of a task exception handling system provided in an embodiment of the present invention includes a training server 101 and an exception handling terminal 102, where the training server 101 and the exception handling terminal 102 are in communication with each other, the training server 101 is configured to run a training task, and the exception handling terminal 102 may monitor the training task running in the training server 101 in real time, so as to execute a task exception handling method provided in an embodiment of the present invention.
The training server and the exception handling terminal can be realized by software, hardware or a combination of software and hardware. For example, the training server may be a server, and the exception handling terminal may be a client computer, or the training server and the exception handling terminal are software, and the training server and the exception handling terminal may be deployed on the same server or client.
The algorithm training system is briefly described below. The algorithm training system is a system for generating a model by training an algorithm, and the algorithm training system can be deployed on at least one server. For example, the algorithm training system may train the CNN algorithm to obtain a face recognition model for face recognition. A simple algorithm training process includes: acquiring training data; processing the training data by using an algorithm to be trained to obtain a processing result; and judging whether the processing result meets the expectation, if not, adjusting the parameters of the algorithm, carrying out next training, and if so, finishing the training of the algorithm to obtain a trained algorithm model. In the embodiment of the present invention, the process of acquiring the training data each time, obtaining the processing result by using the algorithm data, and adjusting the parameters of the algorithm is referred to as a training task.
The algorithm training system may be abnormal during each execution of the training task. For example, an exception due to an environmental problem with the algorithmic training system, an exception due to a problem with the input data of the training task, an exception due to an error in the running of the code logic of the algorithmic training system, and the like. When an abnormality occurs in the running process of the training task, the abnormality occurring in the running process needs to be processed. In the related art, a manual processing mode is mostly adopted, that is, after the training task runs abnormally, a worker needs to be informed to manually process the abnormal running of the training task so as to recover the normal running of the algorithm training system. However, with the scale expansion of the algorithm training system and the increase of training tasks, the frequency of the training tasks having abnormalities in the algorithm training process is higher and higher, and the efficiency is lower only through a manual processing mode.
Illustratively, the training task is a training task for a neural network algorithm, and the algorithm model is a neural network algorithm model.
In order to solve the technical problem of low exception handling efficiency, an embodiment of the present invention provides a task exception handling method, including:
when the target training task in the algorithm training system is detected to be abnormal in operation, acquiring a target abnormal reason causing the abnormal operation of the target training task;
judging whether a preset processing strategy corresponding to the target abnormal reason exists or not;
if the target training task exists, processing the abnormity occurring in the operation of the target training task according to a processing strategy;
and if not, outputting a notification message aiming at the target training task, and storing a processing strategy which is configured by the configuration operation and corresponds to the target abnormal reason when the configuration operation aiming at the target abnormal reason is received.
In the above-mentioned scheme provided by the embodiment of the present invention, when a processing policy corresponding to a target exception cause exists, the exception can be processed according to the processing policy, so that the exception is automatically processed, and the exception processing efficiency is improved.
The task exception handling method according to the embodiment of the present invention will be described in detail below with reference to the accompanying drawings.
As shown in fig. 2, a task exception handling method provided in an embodiment of the present invention may include the following steps:
s201, when the abnormal operation of a target training task in the algorithm training system is detected, acquiring a target abnormal reason causing the abnormal operation of the target training task;
the target training task may be any training task with abnormal operation in the algorithm training system. When the algorithm training system only needs each training task, the running state of each training task in the algorithm training system can be monitored in real time, and when any training task in each training task runs abnormally, the training task running abnormally can be determined as the target training task.
It should be noted that, for the algorithm training system, after any training task runs abnormally, the algorithm training system may generate an abnormal code for the training task with the abnormal operation, and the abnormal code may indicate an abnormal reason of the training task with the abnormal operation, so that the execution subject in the embodiment of the present application may obtain the reason that the target training task runs abnormally by calling a preset interface of the algorithm training system.
Specifically, a task log when the operation of the target training task is abnormal can be obtained through a query interface provided by the algorithm training system. The algorithm training system can generate a task log after the training task is abnormal in operation, wherein the task log records abnormal reasons when the training task is abnormal in operation, such as fault codes, and the like, except for recording the abnormal reasons causing the training task to be abnormal in operation, the task log can also record attribute information of input data of the training task, such as data identification, data format and the like, or can also record abnormal positions when the training task is abnormal in operation, such as the abnormal occurrence in a data acquisition stage, the abnormal occurrence in a data processing stage and the like.
In the embodiment of the present invention, the obtained exception cause may be a target exception cause, and then a subsequent processing step is performed, where the exception cause may be an exception cause identifier, such as a fault code.
S202, judging whether a preset processing strategy corresponding to the target abnormal reason exists or not;
the processing policy corresponding to the abnormal cause may be preset for the abnormal cause, and the processing policy corresponding to the abnormal cause is used to instruct to automatically process the abnormality caused by the abnormal cause. For example, for an exception of training task interruption, the target exception cause is an input error, and the preset processing policy is: the input data is retrieved.
Because there are many exceptions that appear in the training task running process, some exceptions are frequently appeared, and the probability of some exceptions is low, and the exceptions that appear in the training task may be unexpected exceptions, and the exception causes are various, therefore, it is difficult to preset a corresponding processing strategy for each exception cause that causes an exception. Therefore, the exception cause existing in the corresponding processing strategy is often only a part of exception causes, and generally speaking, the exception cause existing in the corresponding processing strategy is often an exception cause which appears more frequently and/or has appeared historically.
Therefore, it is possible to search for whether or not a processing policy corresponding to the target abnormality cause exists from the correspondence relationship between the abnormality cause and the processing policy established in advance. If there is a preset processing strategy corresponding to the target exception cause, step S203 may be executed to automatically process the exception occurring during the operation of the target training task based on the corresponding processing strategy, and if there is no preset processing strategy corresponding to the target exception cause, step S204 may be executed to manually process the exception occurring during the operation of the target training task.
S203, processing the abnormity occurring in the operation of the target training task according to a processing strategy;
in this step, when it is determined that a preset processing strategy corresponding to the target abnormal reason exists, it is indicated that a processing strategy for processing the abnormality occurring during the operation of the target training task is preset, and at this time, the abnormality occurring during the operation of the target training task can be processed according to the processing strategy corresponding to the target abnormal reason, so that when the operation of the target training task is abnormal, the automatic processing of the abnormality is realized.
For example, for an exception of training task interruption, the target exception cause is an input error, and the corresponding processing policy is: if the input data is reacquired, the algorithm training system can be informed to reacquire the input data once.
And S204, outputting a notification message aiming at the target training task, and storing a processing strategy which is configured by the configuration operation and corresponds to the target abnormal reason when the configuration operation aiming at the target abnormal reason is received.
The output notification message for the target training task may be used to indicate that an exception occurring during the operation of the target training task is manually handled.
In this step, when it is determined that there is no preset processing strategy corresponding to the reason of the target anomaly, it is indicated that there is no preset corresponding processing strategy for the target anomaly, and at this time, the anomaly occurring during the operation of the target training task cannot be automatically processed. In order to handle the exception in time, a notification message for the target training task may be output, so as to notify a developer to manually handle the exception occurring during the operation of the target training task.
For example, the notification message may be "target training task exception, please process" to remind the developer to manually process the exception occurring during the operation of the target training task. Of course, besides characters, other forms such as video, pictures, audio, and the like may also be used as the presentation form of the notification message, which is not specifically limited in the embodiment of the present invention. Optionally, the notification manner of the notification message includes any interaction manner such as a short message, an application message, a mail, and a telephone, which is not specifically limited in this embodiment of the present application.
When it is determined that there is no preset processing policy corresponding to the target abnormality cause, the processing policy corresponding to the target abnormality cause configured by the configuration operation may be stored when the configuration operation for the target abnormality cause is received.
The configuration operation may be an operation of configuring the processing strategy by a developer for an abnormal reason causing an abnormality in the operation of the target training task, so that when the abnormality caused by the abnormal reason is encountered again in the following process, the automatic processing according to the configured processing strategy may be performed.
In the above scheme provided by the embodiment of the application, when a processing policy corresponding to a target exception cause exists, the exception can be processed according to the processing policy, so that the exception is automatically processed, and the exception processing efficiency is improved.
Optionally, in an embodiment, the task exception may be automatically handled by using a test case.
The test case is used for indicating and generating a test task for testing the algorithm training system, the test task can be understood as a training task for testing, and the test case generally contains various parameters required by testing, such as formats of input data and the like.
In the embodiment of the invention, when the test cases are used for automatically processing the task exception, the test cases can also contain exception reasons, each test case is correspondingly provided with a case state, and the case state is used for indicating whether the test case contains a processing strategy corresponding to the exception reason. For example, if the case state of a test case is a complete state, the test case further includes a processing policy corresponding to the abnormal reason, whereas if the case state of a test case is an incomplete state, the test case does not include a processing policy corresponding to the abnormal reason.
It is emphasized that the test cases provided by the embodiments of the present invention are different from the test cases in the related art. For the test cases in the related art, each test case only contains various types of parameters required by the test, and indicates the parameters for generating the test task. The test case provided by the application is set for the abnormal reason, at least comprises the abnormal reason, and can also comprise a processing strategy corresponding to the abnormal reason. Of course, the test cases in the embodiment of the present application and the test cases in the related art are different in generation manner and function, and will be described in detail in the following embodiments.
The test case may be stored in a test case library, where the test case library is a database storing test cases, and the test case library may be deployed on the same device as the training server and/or the exception handling terminal, or may be deployed independently of the device where the training server and the exception handling terminal are located.
Illustratively, when the test case is required to be used for testing the algorithm training system, the exception handling terminal may send a test instruction to the training server, instruct the training server to obtain the test case indicated by the test instruction from the test case library, and then the training server executes the test task according to various parameters required by the test included in the obtained test case.
After the target abnormal reason is determined, whether a test case containing a processing strategy corresponding to the target abnormal reason exists or not can be searched based on the target abnormal reason, and whether the processing strategy corresponding to the target abnormal reason exists or not is determined.
Therefore, when there is no preset processing policy corresponding to the target abnormality cause, in order to automatically process the abnormality caused by the target abnormality cause when the abnormality caused by the target abnormality cause is encountered again, a test case including the processing policy corresponding to the target abnormality cause can be generated.
In this case, before outputting the notification message for the target training task, a test case that includes the cause of the target abnormality and has a case status of incomplete may be generated based on the training data of the target training task.
The training data of the target training task may include at least one of input data, an input format, an input position, a trained algorithm model, model parameters when the task is abnormal, environment data of the training task, a stage where the task is abnormal, and the like, and the training environment of the task may include parameters such as a component called by the algorithm model.
After the training data of the target training task is acquired, a test case which contains the reason of the target abnormality and has an incomplete case state can be generated based on the training data of the target training task. Optionally, the test data required for constructing the test task, for example, data input by the target training task, a format of the input data, a trained model, model parameters when the task is abnormal, and the like, may be determined from the training data of the target training task, and further, the cause of the target abnormality is recorded in the test case, and the state of the test case is set to be incomplete, so as to generate a test case that includes the cause of the target abnormality and has an incomplete case state.
For example, the exception handling terminal may read test data required for constructing the test task from the training server terminal, and then write the test data into the test case library according to an agreed storage format of the test case to obtain the test case, and further write the target exception cause into the obtained test case, and further set the state of the test case in the test case library to be incomplete through a state control instruction. The test case library may be deployed in a device where the exception handling end is located, or may also be deployed in a device where the training server end is located.
After generating the test case which includes the target abnormal cause and has an incomplete case state, if a configuration operation for the target abnormal cause is received, a processing policy corresponding to the target abnormal cause and configured by the configuration operation may be set in the generated test case, and the case state of the generated test case may be set to be a complete state.
In the case of implementing automatic processing of task exception by using test case, the determining whether there is a preset processing policy corresponding to a target exception cause may include:
judging whether a first test case which contains a target abnormal reason and is in a finished state exists or not;
optionally, after the target abnormality cause is determined, a first test case which contains the target abnormality cause and is in a complete state may be searched from the test cases contained in the test case library.
In an implementation manner of the determination, a test case in a complete state may be determined from test cases included in a test case library, and the test case may be used as a preliminary screening test case, and then a test case including a target abnormality cause may be searched from the preliminary screening test cases. And if the test case containing the target abnormal reason is found, taking the test case as a first test case, and if the test case containing the target abnormal reason is not found, judging that the first test case does not exist.
In another implementation manner of the determination, a test case including a target abnormal reason may be first searched from each test case included in the test case library, and if the test case including the target abnormal reason is not found, it may be directly determined that the first test case does not exist; otherwise, if the test case containing the target abnormal reason is found, the case state of the found test application is further determined, if the test case is not in the completed state, the first test case is judged to be absent, and if the test case is in the completed state, the found test case is used as the first test case.
After the first test case which contains the reason of the target exception and is in the finished state is found, the exception occurring in the operation of the target training task can be processed according to the processing strategy in the first test case.
Optionally, after the test case which contains the target exception reason and is in the complete state is found, the test case is read from the test case library, and the exception occurring in the operation of the target training task is further processed according to the processing strategy in the test case, so that the exception is automatically processed.
In the above scheme provided by the embodiment of the invention, the test case includes the abnormal reason and can include the processing strategy corresponding to the abnormal reason, so that the processing strategy corresponding to the target abnormal reason can be quickly and accurately obtained by searching the test case, and a realization basis is provided for improving the abnormal processing efficiency.
In one embodiment, each test case further includes: the method is used for verifying whether the processing of the exception caused by the included exception cause is valid, wherein the verification data can be understood as various parameters required for generating the test task, such as the format of the input data, the position of the input data, various model parameters in the runtime and the like.
At this time, in order to ensure the effectiveness of processing the exception occurring during the operation of the target training task, after the exception occurring during the operation of the target training task is processed according to the processing strategy in the first test case, the first test case may be used to verify whether the processing for the exception is effective.
Optionally, the first verification task may be run based on verification data in the first test case, where each verification task is used to verify whether processing for the exception is valid; and if the first verification task runs normally, determining that the abnormal processing of the running of the target training task is completed.
The verification data in the first test case is data used for constructing a target verification task, such as an identifier of input data or indication data, a trained algorithm model, an executed training strategy and the like. If the first verification task runs normally, it indicates that no abnormal reason causing the abnormality exists in the current algorithm training system, and at this time, it can be determined that the processing of the abnormality occurring in the running of the target training task is completed, that is, the processing of the abnormality occurring in the running of the target training task is effective.
In the above scheme provided by the embodiment of the present invention, a target verification task identical to the target training task may be constructed by using a test case including a target abnormality cause, and whether processing for the algorithm training system is effective or not may be verified by using the target verification task.
Optionally, if the first verification task runs abnormally, displaying the abnormality of the target training task during running, so as to manually process the abnormality of the target training task during running.
If the first verification task runs abnormally, it is indicated that an abnormal reason causing the abnormality still exists in the current algorithm training system, that is, the automatic processing of the algorithm training system is invalid, at this time, the abnormality occurring in the running of the target training task can be displayed, for example, prompt information is generated in a display screen to inform a worker to manually process the abnormality which is invalid in the automatic processing.
Optionally, after each manual processing is completed, the step of executing the verification data based on the first test case and running the first verification task may be returned until the first verification task runs normally, that is, until the abnormality occurring during the running of the target training task is effectively processed.
When the first verification task runs normally, it is indicated that the exception caused by the target exception reason is effectively processed, and since the processing policy in the first test case cannot effectively process the exception caused by the target exception reason, the processing policy in the first test case needs to be updated.
Therefore, if an update operation of the processing policy for the target exception cause is received, the processing policy in the first test case can be updated according to the update operation, so that the processing policy in the first test case can effectively process the exception caused by the target exception cause.
Optionally, in another embodiment of the present invention, in a case that each test case further includes verification data for verifying whether processing for an exception caused by the included exception reason is valid, in the generated test case, a processing policy corresponding to a target exception reason and configured by a configuration operation is set, and before a case state of the generated test case is set to a completion state, a second verification task may be executed based on the verification data in the generated test case.
The second verification task is used to verify whether the processing policy configured by the configuration operation can effectively process the exception caused by the target exception reason, and if the second verification task is operating normally, it indicates that the processing policy configured by the configuration operation can effectively process the exception caused by the target exception reason.
Optionally, if the second verification task runs abnormally, it is described that the processing strategy configured by the configuration operation cannot effectively process the abnormality caused by the target abnormality, at this time, the worker needs to be prompted to manually process the abnormality again, that is, the abnormality occurring during the running of the target training task is displayed, the processing of the abnormality occurring during the running of the target training task is prompted, after the processing of the abnormality of the second verification task is completed, the step of executing the verification data based on the generated test case is returned, and the step of running the second verification task is executed until the second verification task runs normally.
In the above scheme provided by the embodiment of the present invention, the verification task can be run through the verification data in the first test data, so that the exception can be effectively handled, and the processing policy recorded in the test case can be effectively exception.
In order to more clearly illustrate the embodiment of the present invention, as shown in fig. 3, the embodiment of the present invention further provides a framework flowchart of task exception handling, which includes a monitoring module, a data collection module, and an automatic verification module.
The monitoring module can detect the training task state of a training task in the algorithm training system in real time, and when the training task fails, an abnormal task training report of the failed task can be generated to provide basic information for subsequent analysis.
The data acquisition module can acquire the related information of the abnormal training task from the training system, acquire the abnormal task training report generated by the monitoring module, analyze the training report, extract the information recorded in the abnormal task training report, further perform preliminary judgment and classification on the failure reason of the failed task, and divide the failed task into tasks capable of being automatically repaired, such as: data format problems, or non-automatic repair tasks, such as: code logic problems. When the data acquisition module analyzes that the failed task is the task which can be automatically repaired, the automatic repair flow can be executed, and when the failed task is analyzed to be the task which can not be automatically repaired, a mail is sent to inform a corresponding developer so as to manually process the abnormity.
The automatic verification module is used for initiating the same training task once again according to the abnormal training task test case acquired by the data acquisition module after the related repair means (automatic repair or manual repair) is executed, so as to verify the repair result after the abnormal repair.
The following is illustrated with reference to examples:
in an example, the target training task is a speech recognition task a for training a speech recognition model, when an error occurs in the operation of the speech recognition task a, it is determined that the target anomaly is caused by loss of input data, and after the execution of the repairing means is completed, a speech recognition task a identical to the speech recognition task a can be automatically generated to verify whether the input data can be successfully acquired.
In the second example, the target training task is a video identification task B for training a video identification model, when an error occurs in the running of the video identification task B, it is determined that the reason of the target abnormality is coding failure, and after the execution of the repair means is completed, a video identification task B identical to the video identification task B can be automatically generated to verify whether coding can be successfully performed.
In the third example, the target training task is a picture classification task C for training a picture classification model, when an error occurs in the operation of the picture classification task C, it is determined that the target abnormality is component loss, and after the execution of the repair means is completed, a picture classification task C identical to the picture classification task C can be automatically generated to verify whether the component is still lost.
And in the fourth example, the target training task is a face recognition task D for training a face recognition model, when the face recognition task D has an error in operation, the target abnormal reason is determined to be the loss of the labeled data, and after the execution of the repairing means is finished, a face recognition task D which is the same as the face recognition task D can be automatically generated to verify whether the labeled data can be successfully obtained.
And in the fifth example, the target training task is a license plate recognition task E for training a license plate recognition model, when the license plate recognition task E has an error in operation, the target abnormality is determined to be caused by the loss of the labeled data, and after the repair means is completed, a license plate recognition task E identical to the license plate recognition task E can be automatically generated so as to verify whether the labeled data can be successfully obtained.
And in the sixth example, the target training task is a motion recognition task F for training a motion recognition model, when an error occurs in the operation of the motion recognition task F, the reason of the target abnormality is determined to be a feature extraction error, and after the execution of the repairing means is completed, the motion recognition task F which is the same as the motion recognition task F can be automatically generated to verify whether the feature can be successfully extracted.
And in the seventh example, the target training task is a foreground and background distinguishing task G for training a foreground and background distinguishing model, when the foreground and background distinguishing task G has an error in operation, the target abnormity reason is determined to be the error in operation of the model, and after the execution of the repairing means is finished, the foreground and background distinguishing task G which is the same as the foreground and background distinguishing task G can be automatically generated so as to verify whether the model can be operated successfully.
After the automatic verification module passes the verification, the data acquisition module collects the information of the current failed task and the solution, and the solution is added into the case library to ensure the richness of the system test cases and the completeness of the automatically repairable scene, so that the case library can be used for carrying out comprehensive test in the development and test stage, and the stability and the robustness of the system are ensured.
In the above scheme provided by the embodiment of the present invention, when a processing policy corresponding to a target exception cause exists, the exception can be processed according to the processing policy, so that automatic processing of the exception is realized, and the exception processing efficiency is improved. Furthermore, the case library can be used for carrying out comprehensive testing in the system development and testing stage, and the stability and the robustness of the system are guaranteed.
Corresponding to the method provided by the foregoing embodiment, as shown in fig. 4, an embodiment of the present invention further provides a task exception handling apparatus, where the apparatus includes:
an abnormal cause obtaining module 401, configured to, when it is detected that a target training task in an algorithm training system runs abnormally, obtain a target abnormal cause that causes the target training task to run abnormally;
a processing policy determining module 402, configured to determine whether a preset processing policy corresponding to the target abnormal cause exists;
a first exception handling module 403, configured to, if the exception exists, handle an exception occurring during operation of the target training task according to the handling policy;
and a second exception handling module 404, configured to output a notification message for the target training task if the exception does not exist, and store a processing policy corresponding to the target exception cause and configured by the configuration operation when the configuration operation for the target exception cause is received.
Optionally, the second exception handling module is further configured to generate, based on the training data of the target training task, a test case that includes the reason for the target exception and has a case status of incomplete, before the notification message for the target training task is output;
the second exception handling module is further configured to set, in the generated test case, a handling policy corresponding to the target exception cause and configured by the configuration operation, and set a case state of the generated test case to a completion state.
Optionally, the processing policy determining module is specifically configured to determine whether a first test case that includes the target exception cause and is in a complete state exists; if so, judging that a preset processing strategy corresponding to the target abnormal reason exists; and if not, judging that a preset processing strategy corresponding to the target abnormal reason does not exist.
Optionally, the first exception handling module is specifically configured to handle an exception occurring during the operation of the target training task according to a handling policy in the first test case.
Optionally, each test case further includes: verification data for verifying whether processing for an exception caused by the included exception cause is valid;
the first exception handling module is further configured to, after handling the exception occurring when the target training task runs according to the handling policy in the first test case, run a first verification task based on verification data in the first test case; wherein each verification task is used for verifying whether the processing aiming at the exception is effective or not; and if the first verification task runs normally, determining that the abnormal processing of the target training task is completed.
Optionally, the first exception handling module is further configured to, if the first verification task runs abnormally, display an exception occurring when the target training task runs, so as to manually handle the exception occurring when the target training task runs; after manual processing, returning to the step of executing the verification data based on the first test case and running a first verification task until the first verification task runs normally; after the first verification task runs normally, if an updating operation of the processing strategy for the target abnormal reason is received, updating the processing strategy in the first test case according to the updating operation.
Optionally, the second exception handling module is further configured to set, in the generated test case, a processing policy corresponding to the target exception cause and configured by the configuration operation, and run a second verification task based on verification data in the generated test case before setting a case state of the generated test case to a completion state;
the second exception handling module is specifically configured to set, in the generated test case, a handling policy corresponding to the target exception cause and configured by the configuration operation if the second verification task runs normally, and set a case state of the generated test case to a completion state.
Optionally, the apparatus further comprises:
the verification module is used for displaying the abnormity of the running of the target training task and prompting to process the abnormity of the running of the target training task if the running of the second verification task is abnormal; and after the exception processing of the second verification task is completed, returning to execute the verification data in the generated test case, and running the second verification task until the second verification task runs normally.
In the above-mentioned scheme provided by the embodiment of the present invention, when a processing policy corresponding to a target exception cause exists, the exception can be processed according to the processing policy, so that the exception is automatically processed, and the exception processing efficiency is improved.
Corresponding to the method provided by the foregoing embodiment, as shown in fig. 1, an embodiment of the present invention further provides a task exception handling system, where the task exception handling system includes: a training server 101 and an exception handling terminal 102;
the training server 101 is used for running a target training task in the algorithm training system;
the exception handling terminal 102 is configured to, when it is detected that the target training task is abnormal in operation, obtain, from the training server 101, a target exception reason causing the target training task to be abnormal in operation; judging whether a preset processing strategy corresponding to the target abnormal reason exists or not; if the target training task exists, controlling the training server according to the processing strategy so as to process the abnormity occurring in the operation of the target training task; and if not, outputting a notification message aiming at the target training task, and storing a processing strategy which is configured by the configuration operation and corresponds to the target abnormal reason when the configuration operation aiming at the target abnormal reason is received.
In the above scheme provided by the embodiment of the present invention, when a processing policy corresponding to a target exception cause exists, the exception can be processed according to the processing policy, so that automatic processing of the exception is realized, and the exception processing efficiency is improved.
An embodiment of the present invention further provides an electronic device, as shown in fig. 5, which includes a processor 501, a communication interface 502, a memory 503 and a communication bus 504, where the processor 501, the communication interface 502 and the memory 503 complete mutual communication through the communication bus 504,
a memory 503 for storing a computer program;
the processor 501 is configured to implement the method steps provided in the above embodiments when executing the program stored in the memory 503.
According to the electronic device provided by the embodiment of the invention, when the processing strategy corresponding to the target exception reason exists, the exception can be processed according to the processing strategy, the exception does not need to be processed manually, and the exception is automatically processed, so that the exception processing efficiency is improved.
The communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the electronic equipment and other equipment.
The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.
In another embodiment of the present invention, a computer-readable storage medium is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any one of the above task exception handling methods.
In another embodiment, the present invention further provides a computer program product containing instructions, which when run on a computer, causes the computer to execute any one of the task exception handling methods in the above embodiments.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the apparatus, device, and system embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference may be made to some descriptions of the method embodiments for relevant points.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. A task exception handling method, the method comprising:
when the abnormal operation of a target training task in an algorithm training system is detected, acquiring a target abnormal reason causing the abnormal operation of the target training task;
judging whether a preset processing strategy corresponding to the target abnormal reason exists or not;
if the target training task exists, processing the abnormity occurring in the operation of the target training task according to the processing strategy;
and if not, outputting a notification message aiming at the target training task, and storing a processing strategy which is configured by the configuration operation and corresponds to the target abnormal reason when the configuration operation aiming at the target abnormal reason is received.
2. The method of claim 1, wherein prior to the outputting the notification message for the target training task, the method further comprises:
generating a test case which contains the target abnormal reason and has an incomplete case state based on the training data of the target training task;
the storing of the processing policy corresponding to the target exception cause configured by the configuration operation includes:
and in the generated test case, setting a processing strategy which is configured by the configuration operation and corresponds to the target abnormal reason, and setting the case state of the generated test case to be a completion state.
3. The method according to claim 2, wherein the determining whether the preset processing strategy corresponding to the target abnormality reason exists comprises
Judging whether a first test case which contains the target abnormal reason and is in a finished state exists or not;
if so, judging that a preset processing strategy corresponding to the target abnormal reason exists;
and if not, judging that a preset processing strategy corresponding to the target abnormal reason does not exist.
4. The method of claim 3, wherein the processing the exception occurring during the operation of the target training task according to the processing strategy comprises:
and processing the abnormity occurring in the operation of the target training task according to the processing strategy in the first test case.
5. The method of claim 4, wherein each test case further comprises: verification data for verifying whether processing for an exception caused by the included exception cause is valid;
after the processing the exception occurring during the operation of the target training task according to the processing strategy in the first test case, the method further includes:
running a first verification task based on verification data in the first test case; wherein each verification task is used for verifying whether the processing aiming at the exception is effective or not;
and if the first verification task runs normally, determining that the abnormal processing of the target training task is completed.
6. The method of claim 5, further comprising:
if the first verification task runs abnormally, displaying the abnormality of the target training task during running, and prompting to process the abnormality of the target training task during running;
after the exception is processed, returning to execute the verification data based on the first test case and running a first verification task until the first verification task runs normally;
after the first verification task runs normally, if an updating operation of the processing strategy for the target abnormal reason is received, updating the processing strategy in the first test case according to the updating operation.
7. The method according to claim 5, wherein before setting, in the generated test case, the processing policy corresponding to the target exception cause and configured by the configuration operation, and setting a case state of the generated test case to a completion state, the method further comprises:
running a second verification task based on verification data in the generated test case;
the setting, in the generated test case, a processing policy corresponding to the target abnormality cause and configured by the configuration operation, and the setting of the case state of the generated test case to a completion state include:
if the second verification task runs normally, setting a processing strategy configured by the configuration operation and corresponding to the target abnormal reason in the generated test case, and setting the case state of the generated test case to be a completion state.
8. The method of claim 7, further comprising:
if the second verification task runs abnormally, displaying the abnormality of the target training task, and prompting to process the abnormality of the target training task;
and after the exception processing of the second verification task is completed, returning to execute the verification data in the generated test case, and running the second verification task until the second verification task runs normally.
9. A task exception handling apparatus, the apparatus comprising:
the abnormal reason acquisition module is used for acquiring a target abnormal reason causing the abnormal operation of the target training task when the abnormal operation of the target training task in the algorithm training system is detected;
the processing strategy judging module is used for judging whether a preset processing strategy corresponding to the target abnormal reason exists or not;
the first exception handling module is used for handling the exception occurring in the operation of the target training task according to the handling strategy if the exception occurs;
and the second exception handling module is used for outputting a notification message aiming at the target training task if the exception does not exist, and storing a handling strategy which is configured by the configuration operation and corresponds to the target exception reason when the configuration operation aiming at the target exception reason is received.
10. A task exception handling system, the task exception handling system comprising: training a server side and an exception handling side;
the training server is used for operating a target training task in the algorithm training system;
the abnormity processing terminal is used for acquiring a target abnormity reason causing the abnormity of the operation of the target training task from the training server terminal when the abnormity of the operation of the target training task is detected; judging whether a preset processing strategy corresponding to the target abnormal reason exists or not; if the target training task exists, controlling the training server according to the processing strategy so as to process the abnormity occurring in the operation of the target training task; and if not, outputting a notification message aiming at the target training task, and storing a processing strategy which is configured by the configuration operation and corresponds to the target abnormal reason when the configuration operation aiming at the target abnormal reason is received.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111308542.1A CN114020432A (en) | 2021-11-05 | 2021-11-05 | Task exception handling method and device and task exception handling system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111308542.1A CN114020432A (en) | 2021-11-05 | 2021-11-05 | Task exception handling method and device and task exception handling system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114020432A true CN114020432A (en) | 2022-02-08 |
Family
ID=80061603
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111308542.1A Pending CN114020432A (en) | 2021-11-05 | 2021-11-05 | Task exception handling method and device and task exception handling system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114020432A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114757278A (en) * | 2022-04-15 | 2022-07-15 | 北京科杰科技有限公司 | Invalid task processing method and device, computer equipment and storage medium |
| CN116541312A (en) * | 2023-07-06 | 2023-08-04 | 广汽埃安新能源汽车股份有限公司 | Continuous integration test method and system for automobile software |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109408320A (en) * | 2018-09-03 | 2019-03-01 | 中国平安人寿保险股份有限公司 | Abnormality eliminating method, device, computer equipment and storage medium are developed in front end |
| CN110750377A (en) * | 2019-09-06 | 2020-02-04 | 深圳平安通信科技有限公司 | Fault positioning method and device |
| CN110990179A (en) * | 2019-11-28 | 2020-04-10 | 中国银行股份有限公司 | Task processing method, device and equipment |
| CN112799804A (en) * | 2021-01-15 | 2021-05-14 | 北京明略软件系统有限公司 | Task management method and system |
| WO2021184554A1 (en) * | 2020-03-18 | 2021-09-23 | 平安科技（深圳）有限公司 | Database exception monitoring method and device, computer device, and storage medium |
-
2021
- 2021-11-05 CN CN202111308542.1A patent/CN114020432A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109408320A (en) * | 2018-09-03 | 2019-03-01 | 中国平安人寿保险股份有限公司 | Abnormality eliminating method, device, computer equipment and storage medium are developed in front end |
| CN110750377A (en) * | 2019-09-06 | 2020-02-04 | 深圳平安通信科技有限公司 | Fault positioning method and device |
| CN110990179A (en) * | 2019-11-28 | 2020-04-10 | 中国银行股份有限公司 | Task processing method, device and equipment |
| WO2021184554A1 (en) * | 2020-03-18 | 2021-09-23 | 平安科技（深圳）有限公司 | Database exception monitoring method and device, computer device, and storage medium |
| CN112799804A (en) * | 2021-01-15 | 2021-05-14 | 北京明略软件系统有限公司 | Task management method and system |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114757278A (en) * | 2022-04-15 | 2022-07-15 | 北京科杰科技有限公司 | Invalid task processing method and device, computer equipment and storage medium |
| CN116541312A (en) * | 2023-07-06 | 2023-08-04 | 广汽埃安新能源汽车股份有限公司 | Continuous integration test method and system for automobile software |
| CN116541312B (en) * | 2023-07-06 | 2023-09-22 | 广汽埃安新能源汽车股份有限公司 | Continuous integration test method and system for automobile software |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109284269B (en) | Abnormal log analysis method and device, storage medium and server | |
| CN106844217B (en) | Method and device for embedding point of applied control and readable storage medium | |
| CN110704304B (en) | Application program testing method and device, storage medium and server | |
| CN117421217B (en) | Automatic software function test method, system, terminal and medium | |
| CN112650688A (en) | Automated regression testing method, associated device and computer program product | |
| CN114020432A (en) | Task exception handling method and device and task exception handling system | |
| CN108153654B (en) | Log collection method and device | |
| CN109542722A (en) | Anomaly analysis processing method, device and storage medium | |
| US20240320082A1 (en) | Systems and methods for pause-correct-replay workflow customization | |
| CN111835566A (en) | System fault management method, device and system | |
| CN111125990A (en) | Method for judging result correctness of parasitic parameters | |
| CN114500249B (en) | Root cause positioning method and device | |
| CN112612882B (en) | Review report generation method, device, equipment and storage medium | |
| CN111198798B (en) | Service stability measuring method and device | |
| CN113037521B (en) | Method for identifying state of communication equipment, communication system and storage medium | |
| CN110413516B (en) | Method and device for identifying slow SQL codes and electronic equipment | |
| TWI778634B (en) | Method for classifying faults, electronic equipment and storage medium | |
| CN114860549B (en) | Buried data verification method, buried data verification device, buried data verification equipment and storage medium | |
| CN114416603B (en) | UI element testing method and device, storage medium and equipment | |
| CN113886165B (en) | Verification method, device and equipment for firmware diagnosis function and readable medium | |
| CN111522737B (en) | Automatic test verification method and device for front-end interface and storage medium | |
| CN112579471B (en) | Method and device for processing software test information | |
| CN117762444A (en) | Software package updating method and device, electronic equipment and storage medium | |
| CN113238963A (en) | Test report generation method, device, equipment, storage medium and program product | |
| CN117827661A (en) | Code grammar checking method, system, computer device and storage medium |
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 |