A terminal testing method, electronic device, and software product

Abstract

This disclosure provides a terminal testing method, electronic device, and program product, relating to the field of testing technology. It solves the problems of manual testing, which consumes significant human resources and wastes algorithm resources, and the need for the algorithm process to be constantly running during automated testing, resulting in high terminal system resource consumption. The terminal testing method includes: inspecting at least one test sample pool through a testing process; in response to detecting the presence of test samples in the target test sample pool, sending a start message to the algorithm process to instruct it to start and perform testing on the test samples. The algorithm process is in a closed state when not in testing mode, and the at least one test sample pool includes the target test sample pool. This embodiment provides an automated testing method that reduces manual resource consumption, decouples the testing process from the algorithm process, and reduces terminal system resource consumption.

Description

Technical Field

[0001] This disclosure relates to the field of testing technology, and in particular to a terminal testing method, electronic device, and program product. Background Technology

[0002] With the development of Internet of Things (IoT) and AI (Artificial Intelligence) technologies, automated testing methods for smart terminals are evolving from manual testing and script testing to multi-process automated testing. The introduction of FTTR (Fiber to the Room) technology has further improved the stability and speed of data transmission, providing a solid foundation for real-time data detection indicators.

[0003] However, current testing solutions either employ manual testing, which requires manually inputting test samples into the terminal device for testing, resulting in huge human resource consumption and wasted algorithm resources; or they employ automated testing, but the algorithm process needs to be running continuously during automated testing, consuming a large amount of system resources. Summary of the Invention

[0004] This disclosure provides a terminal testing method, an electronic device, and a program product.

[0005] In a first aspect, embodiments of this disclosure provide a terminal testing method, applied to a first device for setting up a testing process, the method comprising:

[0006] The testing process is used to inspect at least one preset test sample pool.

[0007] In response to the detection of test samples in the target test sample pool, a start message is sent to the algorithm process to instruct the algorithm process to start and perform test processing on the test samples, wherein the algorithm process is in a closed state when not in a test state, and the at least one test sample pool includes the target test sample pool.

[0008] According to the terminal testing method described in the first aspect, the startup message includes identification information of the target test sample pool, and the identification information is used by the algorithm process to determine the target test sample pool based on the identification information after the algorithm process is started according to the startup message.

[0009] According to the terminal testing method described in the first aspect, the startup message includes the test samples in the target test sample pool, and the test samples are used by the algorithm process to perform test processing based on the test samples after the algorithm process is started according to the startup message.

[0010] According to the terminal testing method described in the first aspect, the method further includes:

[0011] In response to the fact that no test sample is detected in the test sample pool within a preset time period and the algorithm process is in the started state, a shutdown message is sent to the algorithm process to shut down the algorithm process.

[0012] According to the terminal testing method described in the first aspect, the method further includes:

[0013] The obtained test samples are placed into the corresponding test sample pool through the Hardware Abstraction Layer (HAL) interface.

[0014] Secondly, embodiments of this disclosure provide a terminal testing method applied to a second device configured with an algorithm process, the method comprising:

[0015] In response to receiving a start message sent by the test process, the algorithm process is started; the start message is a message sent by the test process to the algorithm process in response to detecting the presence of test samples in the target test sample pool.

[0016] The test sample is processed by the algorithm process after it is started, wherein the algorithm process is in a closed state when not in a test state.

[0017] According to the terminal testing method described in the second aspect, the startup message includes the identification information of the target test sample pool;

[0018] The step of testing the test sample through the algorithm process after startup includes:

[0019] The identification information of the target test sample pool is parsed from the startup message;

[0020] The target test sample pool is determined from at least one test sample pool based on the identification information;

[0021] Obtain the test sample from the target test sample pool;

[0022] The obtained test samples are then subjected to testing processing.

[0023] According to the terminal testing method described in the second aspect, the startup message includes the test sample in the target test sample pool;

[0024] The step of testing the test sample through the algorithm process after startup includes:

[0025] The test sample is parsed from the startup message;

[0026] The parsed test samples are then subjected to testing processing.

[0027] According to the terminal testing method described in the second aspect, the method further includes:

[0028] The inference results obtained through the aforementioned test processing are transmitted to the test interfaces corresponding to each algorithm service, so that the test interfaces of each algorithm service can perform statistical analysis of test-related indicators based on the inference results.

[0029] According to the terminal testing method described in the second aspect, the algorithm process includes a general output interface and individual threads corresponding one-to-one with each of the algorithm services;

[0030] The step of transmitting the inference results obtained through the test processing to the test interfaces corresponding to each algorithm service includes:

[0031] The inference result is transmitted to the target total input interface through the total output interface, so that the target total input interface can identify the inference result and send it to the test interface corresponding to each algorithm service. The target total input interface is the data acquisition interface of the test interface corresponding to each algorithm service.

[0032] According to the terminal testing method described in the second aspect, the algorithm process includes threads corresponding to each of the algorithm services;

[0033] The step of transmitting the inference results obtained through the test processing to the test interfaces corresponding to each algorithm service includes:

[0034] For any of the algorithm services, the inference result of the algorithm service is transmitted to the test interface of the algorithm service through the link between the thread corresponding to the algorithm service and the test interface corresponding to the algorithm service.

[0035] Thirdly, embodiments of this disclosure also provide an electronic device, including:

[0036] One or more processors;

[0037] A memory having stored one or more programs, which, when executed by one or more processors, enable the one or more processors to implement the terminal testing method.

[0038] One or more input / output (I / O) interfaces are connected between the processor and the memory and configured to enable information exchange between the processor and the memory.

[0039] Fourthly, this disclosure also provides a computer program product, which includes a computer program that, when executed by a processor, implements the terminal testing method.

[0040] This embodiment of the scheme involves a testing process that inspects at least one preset test sample pool. In response to the detection of test samples in the target test sample pool, a start message is sent to the algorithm process to instruct it to start and perform testing on the test samples. The algorithm process is in a closed state when not in testing mode, and the at least one test sample pool includes the target test sample pool. This embodiment provides an automated testing method, reducing manual resource consumption, and the decoupling of the testing process from the algorithm process reduces terminal system resource consumption. Attached Figure Description

[0041] In the accompanying drawings of the embodiments disclosed herein:

[0042] Figure 1 A flowchart of a terminal testing method for executing a testing process provided in this embodiment of the disclosure;

[0043] Figure 2 Flowchart of the terminal testing method for algorithm process execution provided in this embodiment of the disclosure;

[0044] Figure 3 A flowchart of a first method for testing test samples through an algorithm process after startup, provided in an embodiment of this disclosure;

[0045] Figure 4 A flowchart of a second method for testing test samples through an algorithm process after startup, provided in an embodiment of this disclosure;

[0046] Figure 5 A schematic diagram illustrating a terminal testing method based on a test process, an algorithm process, a HAL interface, and a test interface, provided in an embodiment of this disclosure;

[0047] Figure 6 This is a block diagram of an electronic device provided in an embodiment of the present disclosure. Detailed Implementation

[0048] To enable those skilled in the art to better understand the technical solutions of this disclosure, the communication-sensing data processing method and computer-readable storage medium provided in the embodiments of this disclosure will be described in detail below with reference to the accompanying drawings.

[0049] The present disclosure will be described more fully below with reference to the accompanying drawings; however, the embodiments shown may be embodied in different forms, and the present disclosure should not be construed as limited to the embodiments set forth below. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will enable those skilled in the art to fully understand the scope of the disclosure.

[0050] The accompanying drawings of the embodiments disclosed herein are provided to further illustrate the embodiments of this disclosure and form part of the specification. They are used together with the detailed embodiments to explain this disclosure and do not constitute a limitation thereof. The above and other features and advantages will become more apparent to those skilled in the art from the description of the detailed embodiments with reference to the accompanying drawings.

[0051] Where there is no conflict, the various embodiments of this disclosure and the features thereof in the embodiments may be combined with each other.

[0052] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the disclosure. The term "and / or" as used in this disclosure includes any and all combinations of one or more of the associated enumerated entries. The singular forms "a" and "the" as used in this disclosure are also intended to include the plural forms, unless the context clearly indicates otherwise. The terms "comprising," "made of," etc., as used in this disclosure specify the presence of the stated feature, integral, step, operation, element, and / or component, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof.

[0053] Unless otherwise specified, all terms used in this disclosure (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and this disclosure, and will not be interpreted as having an idealized or overly formal meaning, unless expressly so defined in this disclosure.

[0054] Current automated testing solutions for AI (artificial intelligence) terminal devices (such as AI cameras, AI smart speakers, etc.) mainly include:

[0055] Manual testing: Testers manually input test samples into terminal devices for testing. This method consumes a lot of human resources. In addition, the relevant algorithm processes or threads will load all resources to wait or process, resulting in a large waste of resources.

[0056] Single-algorithm process testing: The algorithm process loads all resources during initialization and then reads the input stream in real time. Although this reduces the manual cost compared to manual testing, it consumes more system resources of the smart terminal.

[0057] As can be seen from the above, current testing solutions either employ manual testing, which requires manually inputting test samples into the terminal device for testing, resulting in huge consumption of human resources and waste of algorithm resources; or they employ automated testing, but the algorithm process needs to be running continuously during automated testing, resulting in significant consumption of system resources.

[0058] This embodiment of the scheme involves a testing process that inspects at least one preset test sample pool. In response to the detection of test samples in the target test sample pool, a start message is sent to the algorithm process to instruct it to start and perform testing on the test samples. The algorithm process is in a closed state when not in testing mode, and the at least one test sample pool includes the target test sample pool. This embodiment provides an automated testing method, reducing manual resource consumption, and the decoupling of the testing process from the algorithm process reduces terminal system resource consumption.

[0059] The solutions disclosed herein can be applied to any testing scenario, including but not limited to algorithm debugging, integration testing, and functional testing of smart terminal products, and can be widely used by smart terminal testers and related algorithm debugging personnel. The solutions disclosed herein are particularly suitable for automated testing in the field of AI smart terminals, such as automated testing in FTTR (Fiber to the Room) scenarios.

[0060] The embodiments of this disclosure will be described in detail below.

[0061] This disclosure provides a terminal testing method, applied to a first device for setting up a testing process, such as... Figure 1 As shown, it includes steps S11-S12:

[0062] S11. Conduct a routine inspection of at least one preset test sample pool through the testing process.

[0063] S12. In response to detecting the presence of test samples in the target test sample pool, a start message is sent to the algorithm process to instruct the algorithm process to start and perform test processing on the test samples. The algorithm process is in a closed state when not in a test state, and the at least one test sample pool includes the target test sample pool.

[0064] In this embodiment of the disclosure, the first device may include, but is not limited to, any smart terminal, such as an AI smart terminal. The AI ​​smart terminal may include, but is not limited to, one or more AI devices, such as an AI camera, an AI speaker, etc.

[0065] In this embodiment of the disclosure, the operating system of the first device is not limited and can be any operating system that can implement the solution of this embodiment of the disclosure. For example, it can include, but is not limited to, embedded operating systems such as Linux systems, which can support multi-threading and real-time processing.

[0066] In this embodiment of the disclosure, there is no limitation on the network environment supported by the first device; it can be any network environment that supports the solution of this embodiment of the disclosure. For example, the network environment that the first device can support may include, but is not limited to:

[0067] Local Area Network (LAN): Based on FTTR technology, it ensures real-time data transmission of smart terminal devices within the same LAN;

[0068] Wide Area Network: Based on FTTR technology, ensure stable data transmission between smart terminal devices and server devices;

[0069] Network bandwidth: at least 10 Mbps (megabits per second) of uplink bandwidth to support real-time transmission of high-definition video and audio;

[0070] Network latency: less than 100 ms (milliseconds) to ensure the real-time performance of algorithm inference and metric statistics.

[0071] In this embodiment of the disclosure, based on the above-described hardware and software environment configuration, the solution of this embodiment of the disclosure can efficiently and stably perform automated testing of smart terminal products in practical application scenarios. It facilitates the statistical analysis of various algorithm business indicators, and is beneficial for testers and algorithm developers to perform their respective subsequent optimization and iterative optimization without interference, saving a lot of manpower and time costs and ensuring the reliability of test results.

[0072] In this embodiment of the disclosure, the solution is mainly based on a preset test process and an algorithm process. The test process and the algorithm process can be set in the same device or in different devices.

[0073] In this embodiment of the disclosure, the testing process is generally set on the end-side device, while the algorithm process can be set on the end-side or on the server side.

[0074] In this embodiment of the disclosure, the execution scheme of the test process is first introduced.

[0075] In this embodiment of the disclosure, the testing process is mainly responsible for checking whether there are test samples in the test sample pool. If there are, the algorithm process is started and notified to test the test samples in the test sample pool through a preset communication method (such as socket). Otherwise, the algorithm process is closed.

[0076] In this embodiment of the disclosure, in known related technologies, test sample acquisition and test algorithm are not decoupled. The test process performs partial algorithm parsing, or the test process and algorithm process are merged into one process, requiring manual triggering of the algorithm process to start testing. Furthermore, the algorithm process can be quite large, and if it is continuously running (for example, a camera in visual testing would be constantly running), it would consume significant resources. The solution in this embodiment of the disclosure treats the test process as a switch for the algorithm process. The test process patrols the device to check for test samples; the algorithm process is only started if test samples are found.

[0077] In this embodiment of the disclosure, the solution allows the algorithm process to remain shut down when there are no test samples, thereby significantly saving device resources.

[0078] In this embodiment of the disclosure, the test sample acquisition and the test algorithm can be decoupled by the test process inspecting the test sample pool and the algorithm process being responsible for testing the test samples, thereby further saving system resources.

[0079] In this embodiment of the disclosure, if the testing process detects that there are test samples in the target test sample pool, it can send a start message to the algorithm process to notify the algorithm process to start and begin sample testing.

[0080] In this embodiment of the disclosure, the startup message may include, but is not limited to, the identification information of the target test sample pool. The identification information is used by the algorithm process to determine the target test sample pool based on the identification information after the startup message is started.

[0081] In this embodiment of the disclosure, the startup message is used to start the algorithm process on the one hand, and on the other hand, it can be used to pass information about the target test sample pool to the algorithm process. For example, after the test process detects the target test sample pool that stores test samples, it can carry the identification information of the target test sample pool (e.g., encoding information, location information, etc.) in the startup message so that the algorithm process can find the target test sample pool according to the identification information of the target test sample pool and obtain the test sample to be tested from the target test sample pool.

[0082] In this embodiment of the disclosure, the startup message may include, but is not limited to, test samples in the target test sample pool. These test samples are used by the algorithm process to perform test processing based on the test samples after the startup message is started.

[0083] In this embodiment of the disclosure, different test sample pools can be set up for different test samples, such as test samples of different types or different attributes. Each test sample is placed in the corresponding test sample pool. Based on the classification of test sample pools, it is easier for the algorithm process to obtain the corresponding test sample from the target test sample pool. Compared with setting up only one test sample pool, this can greatly improve the testing efficiency.

[0084] In this embodiment of the disclosure, the test sample may include, but is not limited to, video samples, audio samples, image samples, etc., and correspondingly, the test sample pool may include, but is not limited to, video test sample pool, audio test sample pool, image test sample pool, etc.

[0085] In this embodiment of the disclosure, the startup message is used to start the algorithm process on the one hand, and to pass test samples to the algorithm process on the other hand. For example, after the test process detects the target test sample pool that stores test samples, it can obtain test samples from the target test sample pool and carry the test samples in the startup message so that the algorithm process can directly test the test samples.

[0086] In this embodiment of the disclosure, the method may further include:

[0087] If no test sample is detected in the test sample pool within a preset time period and the algorithm process is in the running state, a shutdown message is sent to the algorithm process to shut it down.

[0088] In this embodiment of the disclosure, in order to avoid the algorithm process consuming system resources by remaining in the startup state for a long time, the algorithm process in the startup state can be shut down if no test sample is detected within a preset time period.

[0089] In this embodiment of the disclosure, the method may further include:

[0090] The obtained test samples are placed into the corresponding test sample pool through the Hardware Abstraction Layer (HAL) interface.

[0091] In this embodiment of the disclosure, a HAL (Hardware Abstraction Layer) can be abstracted from the hardware interface of a smart terminal, and test samples obtained from one or more smart terminals (such as AI cameras, AI speakers, etc.) can be placed into a test sample pool through the HAL interface.

[0092] In this embodiment of the disclosure, the HAL interface needs to be set in the end device. When the HAL interface is used, it can be independently compiled into an SO (Shared Object) dynamic library.

[0093] This disclosure provides a terminal testing method applied to a second device configured with an algorithm process, such as... Figure 2 As shown, it includes steps S21-S22:

[0094] S21. In response to receiving the start message sent by the test process, start the algorithm process; the start message is a message sent by the test process to the algorithm process in response to detecting the existence of test samples in the target test sample pool.

[0095] S22. The test sample is processed by the algorithm process after startup. The algorithm process is in a closed state when not in test mode.

[0096] In this embodiment of the disclosure, the algorithm process can be responsible for loading and unloading various algorithm services.

[0097] In this embodiment of the disclosure, the algorithm process can remain shut down when there are no test samples to conserve system resources. The algorithm process can be started based on a start message sent by the test process after it detects the target test sample pool.

[0098] In this embodiment of the disclosure, the startup message may include identification information of the target test sample pool.

[0099] In the embodiments disclosed herein, such as Figure 3 As shown, the test sample is processed by the algorithm process after startup, which may include steps S31-S34:

[0100] S31. Parse the identification information of the target test sample pool from the startup message.

[0101] S32. Determine the target test sample pool from at least one test sample pool based on the identification information.

[0102] S33. Obtain test samples from the target test sample pool.

[0103] S34. Perform test processing on the obtained test samples.

[0104] In this embodiment of the disclosure, the algorithm process may contain a message parsing thread, which parses the startup message to obtain the content carried by the startup message.

[0105] In this embodiment of the disclosure, the algorithm process may further include a sample acquisition thread. When the startup message is parsed to contain the identification information of the target test sample pool, the sample acquisition thread can acquire test samples from the target test sample pool based on the identification information of the target test sample pool.

[0106] In this embodiment of the disclosure, the algorithm process may also include a test sample parsing thread and other algorithm threads. The test sample parsing thread can parse the collected test samples. Based on the parsing result (e.g., the algorithm business type of the parsed test sample), the algorithm thread corresponding to the algorithm business can be started to infer the test sample, thereby realizing the test processing of the test sample.

[0107] In this embodiment of the disclosure, the startup message may include test samples from the target test sample pool.

[0108] In the embodiments disclosed herein, such as Figure 4 As shown, the test sample is processed by the algorithm process after startup, which may include steps S41-S42:

[0109] S41. Parse the test sample from the startup message.

[0110] S42. Perform test processing on the parsed test samples.

[0111] In this embodiment of the disclosure, the algorithm process may contain a message parsing thread, which parses the startup message to obtain the content carried by the startup message.

[0112] In this embodiment of the disclosure, the algorithm process may also include a test sample parsing thread and other algorithm threads. When the startup message is parsed and carries a test sample, the test sample parsing thread can parse the parsed test sample. Based on the parsing result (e.g., the algorithm business type of the parsed test sample), the corresponding algorithm thread can be started to infer the test sample, thereby realizing the test processing of the test sample.

[0113] In this embodiment of the disclosure, the method may further include:

[0114] The inference results obtained after testing are sent to the test interfaces corresponding to each algorithm business, so that the test interfaces of each algorithm business can perform statistical analysis of test-related indicators based on the inference results.

[0115] In this embodiment, considering that the algorithm process needs to record the processing strategies of relevant indicators for each algorithm business, which is labor-intensive and tedious, and also consumes system resources, the solution of this embodiment decouples the test interface from the algorithm process. This allows one algorithm process to correspond to multiple test interfaces, meaning that each algorithm business can share the algorithm model in one algorithm process for inference. Multiple threads corresponding to different algorithm businesses can run within the algorithm process. Each algorithm business can have a corresponding algorithm thread and test interface. Each algorithm thread can call the algorithm model in the algorithm process when processing test samples; for example, it can call the same algorithm model.

[0116] In this embodiment of the disclosure, the above-described scheme decouples the test interfaces of each algorithm service from the algorithm process. The inference results are transmitted to the test interface of the algorithm service via a preset communication method (e.g., socket). The test interface of the algorithm service then performs statistical analysis on the test-related metrics according to its own defined evaluation methods or logic. This decoupling of the test interface from the algorithm process ensures the real-time performance of each algorithm thread and the accuracy of the test results, while also enhancing the maintainability of the automated testing system.

[0117] In this embodiment of the disclosure, the test interface corresponding to each algorithm service is an independent part of each algorithm service, which can be responsible for calculating and statistically analyzing the relevant indicators of the algorithm service.

[0118] In this embodiment of the disclosure, the test interface is generally set on the server side, but it can also be set on the client side.

[0119] In this embodiment of the disclosure, considering that placing both the algorithm process and the test interface on the device side would increase the burden on the device side, the placement of the algorithm process and the test interface can be determined based on the computing power of the device side.

[0120] In this embodiment of the disclosure, the test interface can be an actual interface or a logical interface. Its essence is to perform statistics and processing of the indicator data of the recognition results of the algorithm model in the algorithm process, rather than an interface for calling test scripts to perform tests.

[0121] In this embodiment of the disclosure, the algorithm process includes a general output interface and threads corresponding one-to-one with each algorithm service; transmitting the inference results obtained after test processing to the test interface corresponding to each algorithm service may include:

[0122] The inference results are transmitted to the target input interface through the total output interface, so that the target input interface can identify the inference results and send them to the test interfaces corresponding to each algorithm service. The target input interface is the data acquisition interface of the test interface corresponding to each algorithm service.

[0123] In this embodiment of the disclosure, the algorithm process can be configured to include a unified total output interface, and the test interface has one or more total access interfaces (such as the target total access interface mentioned above). The total access interface of the test interface parses the inference result of the algorithm process to determine the test interface corresponding to the algorithm service corresponding to the inference result, and then activates the corresponding test interface to send the inference result to the corresponding test interface for statistical analysis of test-related indicators.

[0124] In this embodiment of the disclosure, the algorithm process may include threads corresponding to each algorithm service; transmitting the inference results obtained after test processing to the test interface corresponding to each algorithm service may include:

[0125] For any algorithm business in each algorithm business, the inference result of the algorithm business is transmitted to the test interface of the algorithm business through the link between the thread corresponding to the algorithm business and the test interface corresponding to the algorithm business.

[0126] In this embodiment of the disclosure, the transmission link can also be set up one-to-one during the algorithm process. That is, a one-to-one transmission link is configured between the algorithm thread and the test interface corresponding to the algorithm thread. After the algorithm thread of the algorithm process obtains the inference result, it can directly transmit the inference result to the corresponding test interface. In this embodiment, all test interfaces need to be activated, otherwise packets cannot be received.

[0127] In this embodiment of the disclosure, a schematic diagram of the terminal testing method based on the test process, algorithm process, test sample pool, HAL interface, and test interface can be shown as follows: Figure 5 As shown: The testing process continuously checks at least one test sample pool for the existence of test samples. The HAL interface reads test samples (such as video samples, voice samples, image samples, etc.) from the smart terminal and places them into the test sample pool. If the testing process detects the existence of a test sample in any of the at least one target test sample pools, it sends a start message to the algorithm process. The algorithm process starts according to the start message and obtains the content carried by the start message, such as the identification information of the target test sample pool or the test sample. If it carries identification information, it retrieves the test sample from the target test sample pool based on the identification information and performs test processing on the test sample. If it carries a test sample, it directly performs test processing on the test sample. After the algorithm process obtains the inference result about the test sample based on the test processing, it sends the inference result to the test interface corresponding to the algorithm thread corresponding to the inference result. The test interface performs statistical analysis of test-related indicators based on the inference result and writes the statistical results to the result file.

[0128] This disclosure embodiment includes at least the following advantages:

[0129] 1. The HAL layer and test sample resource pool are abstracted in the hardware interface to enhance the universality of automated testing methods.

[0130] 2. By inspecting the test sample pool during the testing process, resource utilization is improved and resource waste is avoided.

[0131] 3. The algorithm process is decoupled from the test interfaces of each algorithm business, so that the test interfaces of each algorithm business can effectively use the inference results of the corresponding algorithm thread in the algorithm process to perform statistical analysis of test-related indicators, thereby enhancing the reliability of indicator data.

[0132] 4. The algorithm process remains shut down when there are no test samples, saving system resources.

[0133] 5. The testing process is decoupled from the algorithm process, reducing the complexity of the testing process and further saving system resources.

[0134] This disclosure also provides a first device equipped with a testing process; the testing process is used for:

[0135] Perform a routine inspection of at least one pre-defined test sample pool;

[0136] In response to the detection of test samples in the target test sample pool, a start message is sent to the algorithm process to instruct the algorithm process to start and perform test processing on the test samples. The algorithm process is in a closed state when not in a test state, and the at least one test sample pool includes the target test sample pool.

[0137] This disclosure also provides a second device equipped with an algorithm process; the algorithm process is used for:

[0138] The test process initiates a startup in response to receiving a startup message from the test process. The startup message is sent by the test process to the algorithm process in response to the detection of test samples in the target test sample pool.

[0139] After startup, the test samples are processed for testing. The algorithm process is in a closed state when not in testing mode.

[0140] In the embodiments disclosed herein, any of the embodiments in the foregoing terminal testing method embodiments are applicable to the first device and the second device embodiments, and will not be described in detail here.

[0141] This disclosure also provides an electronic device 100, such as... Figure 6 As shown, it includes:

[0142] One or more processors 101;

[0143] The memory 102 stores one or more programs, which, when executed by the one or more processors, cause the one or more processors 101 to implement the terminal testing method.

[0144] One or more input / output (I / O) interfaces 103 are connected between the processor 101 and the memory 102 and configured to enable information interaction between the processor 101 and the memory 102.

[0145] This disclosure also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the terminal testing method described above.

[0146] This disclosure also provides a computer program product, which includes a computer program that, when executed by a processor, implements the terminal testing method described above.

[0147] In the embodiments disclosed herein, any of the aforementioned terminal testing methods are applicable to the embodiments of the electronic device, storage medium, and program product, and will not be described in detail here.

[0148] Those skilled in the art will understand that all or some of the functional modules / units disclosed above can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0149] In hardware implementations, the division between functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be executed by several physical components working together.

[0150] Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit (CPU), digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technique for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH) or other disk storage; read-only optical disc (CD-ROM), digital versatile disc (DVD) or other optical disc storage; magnetic cartridges, magnetic tapes, disk storage or other magnetic storage; and any other media that can be used to store desired information and can be accessed by a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0151] This disclosure has disclosed exemplary embodiments, and although specific terminology has been used, it is for general illustrative purposes only and should not be construed as limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of this disclosure as set forth by the appended claims.

Claims

1. A terminal testing method, characterized in that, The method, applied to a first device for setting up a test process, includes: The testing process is used to inspect at least one preset test sample pool. In response to the detection of test samples in the target test sample pool, a start message is sent to the algorithm process to instruct the algorithm process to start and perform test processing on the test samples. The inference results obtained after the test processing are transmitted to the test interfaces corresponding to each algorithm service, so that the test interfaces of each algorithm service can perform statistical analysis of test-related indicators based on the inference results. The algorithm process is in a closed state when not in test mode, and the at least one test sample pool includes the target test sample pool. The method further includes: In response to the fact that no test sample is detected in the test sample pool within a preset time period and the algorithm process is in the running state, a shutdown message is sent to the algorithm process to shut down the algorithm process; The obtained test samples are placed into the corresponding test sample pool through the Hardware Abstraction Layer (HAL) interface.

2. The terminal testing method according to claim 1, characterized in that, The startup message includes identification information of the target test sample pool. The identification information is used by the algorithm process to determine the target test sample pool based on the identification information after the algorithm process starts according to the startup message.

3. The terminal testing method according to claim 1, characterized in that, The startup message includes the test samples in the target test sample pool. The test samples are used by the algorithm process to perform test processing based on the test samples after the algorithm process is started according to the startup message.

4. A terminal testing method, characterized in that, The method, applied to a second device configured with an algorithm process, includes: In response to receiving a start message from the test process, the algorithm process is started; the start message is a message sent by the test process to the algorithm process in response to detecting the existence of test samples in the target test sample pool; the test samples are obtained through the Hardware Abstraction Layer (HAL) interface and placed into the corresponding test sample pool. The algorithm process, once started, performs testing on the test samples and transmits the inference results obtained through the testing to the test interfaces corresponding to each algorithm service. Based on the inference results, the test interfaces of each algorithm service then perform statistical analysis of relevant test metrics. The algorithm process is in a closed state when not in testing mode. If the test process does not detect any test samples in the test sample pool within a preset time period, the algorithm process is in a started state. The algorithm process receives a shutdown message from the test process to enter the closed state.

5. The terminal testing method according to claim 4, characterized in that, The startup message includes the identification information of the target test sample pool; The step of testing the test sample through the algorithm process after startup includes: The identification information of the target test sample pool is parsed from the startup message; The target test sample pool is determined from at least one test sample pool based on the identification information; Obtain the test sample from the target test sample pool; The obtained test samples are then subjected to testing processing.

6. The terminal testing method according to claim 4, characterized in that, The startup message includes the test samples in the target test sample pool; The step of testing the test sample through the algorithm process after startup includes: The test sample is parsed from the startup message; The parsed test samples are then subjected to testing processing.

7. The terminal testing method according to claim 4, characterized in that, The algorithm process includes a main output interface and individual threads corresponding to each of the algorithm's business functions. The step of transmitting the inference results obtained through the test processing to the test interfaces corresponding to each algorithm service includes: The inference result is transmitted to the target total input interface through the total output interface, so that the target total input interface can identify the inference result and send it to the test interface corresponding to each algorithm service. The target total input interface is the data acquisition interface of the test interface corresponding to each algorithm service.

8. The terminal testing method according to claim 4, characterized in that, The algorithm process includes threads corresponding to each of the algorithm's business functions; The step of transmitting the inference results obtained through the test processing to the test interfaces corresponding to each algorithm service includes: For any of the algorithm services, the inference result of the algorithm service is transmitted to the test interface of the algorithm service through the link between the thread corresponding to the algorithm service and the test interface corresponding to the algorithm service.

9. An electronic device, comprising: One or more processors; A memory having one or more programs stored thereon, which, when executed by one or more processors, cause the one or more processors to implement the terminal testing method according to any one of claims 1-3 or any one of claims 4-8; One or more input / output (I / O) interfaces are connected between the processor and the memory and configured to enable information exchange between the processor and the memory.

10. A computer program product comprising a computer program that, when executed by a processor, implements the terminal testing method according to any one of claims 1-3 or any one of claims 4-8.

Images