An interpretable road network selection method, device, medium and equipment

Abstract

The present application discloses an interpretable road network selection method, device, storage medium and electronic device. The method includes: obtaining the description features of each test road and constructing a feature vector based on the description features; determining a teacher model and sorting the feature importance of the description features input into the teacher model through an interpretable tool; constructing an additive neural network; calculating a knowledge distillation loss function based on the prediction results of the teacher model; using the sorted description features as a training set and inputting them into the additive neural network, training the additive neural network through the knowledge distillation loss function, and using the trained additive neural network as an interpretable road network selection model to perform road selection through the interpretable road network selection model. The present application can take into account the high accuracy of the road selection model and make it interpretable.

Description

Technical Field

[0001] This application relates to the technical field of geographic information processing, and particularly to an interpretable road network selection method, device, storage medium, and electronic device. Background Art

[0002] Map generalization refers to deriving map data with a smaller scale and more general expression from map data with a larger scale and more detailed expression, so as to construct map products with different scales, such as topographic maps with different scales and network navigation maps at different levels. As a basic geographic element, the road network is one of the main objects of map generalization processing. The map generalization of road network data is mainly reflected in the selection of road targets, that is, selecting some important road targets from the original road data to meet the clarity requirements of the smaller scale expression. Among them, the number of selected road targets can be calculated using the square root model. However, which specific road targets to select is determined by the importance of each road, and multiple factors such as road attribute level, topological connectivity, and geometric length need to be considered overall, which is a complex multi-feature analysis and decision-making problem.

[0003] Under the existing technology system, the selection process of the road network still requires in-depth participation of professional cartographers, with high input and long cycles, which severely restricts the iterative update of map products. In recent years, researchers have introduced machine learning models to develop intelligent road network selection methods to reduce or even replace manual participation in map generalization tasks of the road network. Such methods usually use different descriptive features of road targets as inputs, and use machine learning models to construct classifiers, so as to output selection decision results. Among them, the decision-making knowledge of the classifier is mainly obtained through supervised learning of sample case data. In practical applications, although this machine learning-based selection method can achieve a high accuracy rate, the models used are usually relatively complex in structure, similar to "black boxes", resulting in poor interpretability and difficulty in understanding the internal decision-making logic and rules of the models. Poor model interpretability will further lead to the inability to improve the model performance according to requirements during the model usage process, as well as poor application effects of the models and low user trust. Summary of the Invention

[0004] The embodiments of this application provide an interpretable road network selection method, device, storage medium, and electronic device, which can improve the accuracy of road network selection, and the interpretability of the road selection model improves user trust.

[0005] The embodiments of this application provide an interpretable road network selection method, including:

[0006] Obtain the descriptive features of each test road, and construct a feature vector based on the descriptive features;

[0007] Determine the teacher model, and sort the feature importance of the described features input to the teacher model through an interpretable tool;

[0008] Construct an additive neural network;

[0009] Calculate the knowledge distillation loss function based on the prediction results of the teacher model;

[0010] Take the sorted described features as the training set and input them into the additive neural network. Train the additive neural network through the knowledge distillation loss function, and use the trained additive neural network as an interpretable road network selection model to select roads through the interpretable road network selection model.

[0011] Furthermore, for the above-mentioned interpretable road network selection method, wherein, the determination of the teacher model includes:

[0012] Take the feature vector as the training set and the test set, train multiple different machine learning models through the training set, and input the test set into the trained multiple different machine learning models to obtain prediction results. Select the machine learning model with the highest prediction accuracy as the teacher model.

[0013] Furthermore, for the above-mentioned interpretable road network selection method, wherein, the sorting of the feature importance of the described features input to the teacher model through an interpretable tool includes:

[0014] Calculate the average contribution value of the j -th feature for all test examples:

[0015]

[0016] wherein, represents the total number of test roads, represents the SHAP contribution value of the i -th feature of the j -th target test road;

[0017] Sort the described features in descending order according to the SHAP contribution value.

[0018] Furthermore, for the above-mentioned interpretable road network selection method, wherein, the additive neural network includes an input layer, an intermediate layer and an output layer, and the output of the additive neural network is used to judge whether the test road is selected and retained;

[0019] The mathematical expression of the additive neural network is:

[0020]

[0021] Among them, represents the feature neural network, n represents the number of feature networks, and its maximum value is the total number of the described features.

[0022] Furthermore, for the above-mentioned interpretable road network selection method, wherein calculating the knowledge distillation loss function based on the prediction result of the teacher model includes:

[0023] Using the teacher model to predict the first probability of each test road being selected and retained, and marking the first probability as the first label;

[0024] Calculating the soft label of the test road being selected and retained required for knowledge distillation based on the first label;

[0025] Calculating the knowledge distillation loss function based on the soft label.

[0026] Furthermore, for the above-mentioned interpretable road network selection method, wherein the method further includes:

[0027] Calculating the soft label through a first formula, and the first formula is:

[0028]

[0029] Among them, is the first label, is the distillation temperature.

[0030] Calculating the knowledge distillation loss function through a second formula, and the second formula is:

[0031]

[0032] Among them, is the prediction result of the additive neural network, is the artificial label of the test road, is the weight parameter of the soft label loss and the true label loss, is the mean square error function.

[0033] Furthermore, for the above-mentioned interpretable road network selection method, wherein the described features include road description features and road connectivity features, the road description features at least include road grade, number of lanes, road surface material, length, width, and number of arc segments, and the road connectivity features at least include degree centrality, clustering coefficient, betweenness centrality, closeness centrality, and eigenvector centrality.

[0034] The embodiment of the present application also provides an interpretable road network selection device, including:

[0035] An acquisition module, configured to acquire the descriptive features of each test road and construct a feature vector based on the descriptive features;

[0036] A determination and sorting module, configured to determine a teacher model and sort the feature importance of the descriptive features input into the teacher model through an interpretability tool;

[0037] A construction module, configured to construct an additive neural network for the network;

[0038] A calculation module, configured to calculate a knowledge distillation loss function based on the prediction result of the teacher model;

[0039] A road selection module, configured to use the sorted descriptive features as a training set and input them into the additive neural network, train the additive neural network through the knowledge distillation loss function, use the trained additive neural network as an interpretable road network selection model, and perform road selection through the interpretable road network selection model.

[0040] An embodiment of the present application further provides a computer-readable storage medium, in which multiple instructions are stored, and the instructions are suitable for being loaded by a processor to execute any one of the above-mentioned interpretable road network selection methods.

[0041] An embodiment of the present application further provides an electronic device, including a processor and a memory, the processor is electrically connected to the memory, the memory is used to store instructions and data, and the processor is used for the steps in any one of the above-mentioned interpretable road network selection methods.

[0042] The interpretable road network selection method, device, storage medium and electronic device provided by the present application. The present application trains a student model (additive neural network) by constructing a teacher model, according to the prediction result of the teacher model, and a knowledge distillation loss function, and obtains an interpretable road selection model. On the one hand, the additive neural network itself has interpretability between the input features and the output results. On the other hand, the knowledge distillation method is used to endow the additive neural network with the decision-making knowledge of the teacher model with high accuracy, so that the road selection model takes into account both high accuracy and interpretability. By performing road selection through the road selection model, the accuracy of road network selection can be improved, and the user trust is enhanced due to the interpretability of the road selection model. Description of the Drawings

[0043] The following will make the technical solutions and other beneficial effects of the present application obvious by describing the specific embodiments of the present application in detail in conjunction with the drawings.

[0044] Figure 1 It is a flowchart of the interpretable road network selection method provided by an embodiment of the present application.

[0045] Figure 2 It is a schematic diagram of the result of sorting the importance of features for describing features provided by an embodiment of the present application.

[0046] Figure 3 It is a schematic diagram of the structure of an additive neural network provided by an embodiment of the present application.

[0047] Figure 4 It is a flowchart of training a model by a knowledge distillation method provided by an embodiment of the present application.

[0048] Figure 5 It is a relationship curve graph between feature description and contribution value provided by an embodiment of the present application.

[0049] Figure 6 It is a schematic diagram of the structure of an interpretable road network selection device provided by an embodiment of the present application.

[0050] Figure 7 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application.

[0051] Figure 8 It is another schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0052] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present application.

[0053] An embodiment of the present application provides an interpretable road network selection method, device, storage medium, and electronic device. An interpretable road network selection device provided by an embodiment of the present application can be integrated in an electronic device, and the electronic device can be a device such as a terminal or a server. Among them, the terminal can include a tablet computer, a notebook computer, a personal computer (PC), a microprocessing box, or other devices, etc.

[0054] Please refer to Figure 1 , Figure 1 It is a flowchart of an interpretable road network selection method provided by an embodiment of the present application, which is applied to an electronic device. The interpretable road network selection method includes the following steps:

[0055] S1, obtain the description features of each test road, and construct a feature vector based on the description features.

[0056] First, given a road network in a certain area , denote the test road, m denote the number of test roads included in the road network. Among them, the test road is composed of connected road arcs with the same name.

[0057] Then, obtain the descriptive features of each test road and construct a feature vector, denoted as , K denote the number of features. The road descriptive features adopted include two aspects: one is the road descriptive features, including road grade, number of lanes, road surface material, length, width, and the number of arc segments, etc., and the other is the connectivity features of each road in the entire road network, which can be defined and calculated based on the road network graph model (i.e., road targets are used as graph nodes, and the connection relationships between road targets are expressed as edges), including degree centrality, clustering coefficient, betweenness centrality, closeness centrality, and eigenvector centrality, etc. Then, standardize the feature vector.

[0058] S2. Determine the teacher model, and sort the feature importance of the descriptive features input into the teacher model through an interpretable tool.

[0059] In one embodiment, step S2 includes the following steps:

[0060] S21. Use the feature vector as the training set and the test set, train multiple different machine learning models through the training set, and input the test set into the multiple different trained machine learning models to obtain prediction results, and select the machine learning model with the highest prediction accuracy as the teacher model.

[0061] Specifically, manually label the selection label ("retain" or "discard") of each road, and divide the training set and the test set according to a certain ratio (such as 8:2). Input the training set into different machine learning models respectively to obtain the prediction results of different machine learning models, and perform iterative training according to the labels. Then input the test set into the different trained machine learning models respectively for prediction to obtain the prediction results, and select the machine learning model with the highest prediction accuracy as the teacher model.

[0062] As an example, the machine learning model can select classifier models such as random forest, gradient boosting tree, and BP neural network.

[0063] S22. Calculate the average contribution value of the j th feature for all test examples:

[0064]

[0065] Among them, represents the total number of test roads, represents the i th j SHAP contribution value of the

[0066] Among them, the SHAP contribution value is calculated by the SHAP (SHapley Additive exPlanations) explainable tool.

[0067] S23. Sort the description features in descending order of the SHAP contribution value.

[0068] Figure 2 is a schematic diagram of the feature importance ranking result of the description features provided by the embodiment of the present application. As Figure 2 shown, the ranking results are in turn rank, number of arcs, degree centrality, length, betweenness centrality, closeness centrality, eigenvector centrality, width, clustering coefficient, material, number of lanes. S3. Construct an additive neural network.

[0069] Figure 3 is a schematic diagram of the structure of the additive neural network provided by the embodiment of the present application. As Figure 3 shown, the additive neural network NAM (Neural Additive Model) includes multiple feature neural networks. Each feature neural network includes an input layer, an intermediate layer, and an output layer. The input layer is a single neuron that controls the input of road features. The intermediate layer contains a few neurons and uses the Relu activation function to ensure its non-linear fitting ability. The output layer is a single neuron and uses the Sigmoid activation function. Each feature neural network only accepts a single feature input and outputs the contribution value of this feature , and . The additive neural network is used to judge whether a test road is selected and retained. Denote the additive neural network containing n feature neural networks as . Input the n description features of a certain test road into , and output the prediction result of this test road, that is, = . In practical applications, it can be set that when , it means that the road target is selected and retained.

[0070] The mathematical expression of the additive neural network is:

[0071]

[0072] Among them, represents the feature neural network, nrepresents the number of feature networks, and its maximum value is the total number K of described features.

[0073] S4. Calculate the knowledge distillation loss function based on the prediction results of the teacher model.

[0074] Figure 4 is the flowchart of training a model by the knowledge distillation method provided by the embodiments of the present application, and can be referred to Figure 4 , and step S4 includes the following steps:

[0075] S41. Use the teacher model to predict the first probability of being selected and retained for each test road, and mark the first probability as the first label.

[0076] S42. Calculate the soft label of the test road to be selected and retained required for knowledge distillation based on the first label.

[0077] Specifically, input the sorted described features into the trained teacher model to obtain the prediction results of the teacher model, that is, the probability of being predicted as selected and retained for each test road. Obtain the first probability that the teacher model predicts as selected and retained for each test road and record it as the first label , and calculate the soft label of the test road predicted as selected and retained required for knowledge distillation by the following formula , where T is the distillation temperature, and the larger the value of T, the smoother the predicted probability distribution of different road targets in

[0078]

[0079] Among them, is the first label, is the distillation temperature.

[0080] S43. Calculate the knowledge distillation loss function based on the soft label.

[0081] The knowledge distillation loss is composed of the weighted sum of the hard label (manual label) loss and the soft label loss, as shown in the following formula:

[0082]

[0083] Among them, is the prediction result of the additive neural network, is the manual label of the test road, which is a hard-coded 0 or 1, is the weight parameter of the soft label loss and the true label loss, is the mean square error function.

[0084] The setting of the Loss function enables the predicted output of the student model to approximate the true label while getting as close as possible to the prediction of the teacher model, thereby transferring the decision-making knowledge possessed by the teacher model to the student model.

[0085] S5. Use the sorted descriptive features as the training set and input them into the additive neural network. Train the additive neural network through the knowledge distillation loss function, and use the trained additive neural network as the interpretable road network selection model to perform road selection through the interpretable road network selection model.

[0086] Specifically, construct different combinations of feature quantities according to the sorting of the descriptive features, and input different combinations of feature quantities into the additive neural network model (denoted as ( )). Combine the selected teacher model and loss function Train through the knowledge distillation method, and select the model with the best performance on the test data set as the final interpretable road network selection model.

[0087] It should be noted that different combinations of feature quantities refer to the combinations of the top n important features in the feature sorting. For example, combination 1 is [grade], combination 2 is [grade, number of arcs], and combination 3 is [grade, number of arcs, degree centrality].

[0088] The road network selection model obtained through the above method is interpretable. Figure 5 This is the relationship curve graph between the feature description and the contribution value provided by the embodiment of the present application. As Figure 5 shown, the final interpretable road network selection model is , that is, this model utilizes the 4 most important features in the feature sorting: grade, number of arc segments, degree centrality, and length, and achieves its optimal accuracy. We visualize the curve of 4 of these feature networks: the horizontal axis is the size after normalizing the feature value, the vertical axis is the feature contribution, and the blue dots represent the overall distribution of road target features. Then, the interpretable contribution value of the feature of any 1 road target can be found, that is, as Figure 5 shown by the green dots in.

[0089] The interpretable road network selection method provided by this application trains a student model (additive neural network) by constructing a teacher model and according to the prediction results of the teacher model and the knowledge distillation loss function, and obtains an interpretable road selection model. On the one hand, the additive neural network itself has interpretability between input features and output results. On the other hand, the knowledge distillation method is used to endow the additive neural network with the decision-making knowledge of the teacher model with high accuracy, so that the road selection model takes into account both high accuracy and interpretability. By using the road selection model for road selection, the accuracy of road network selection can be taken into account, and the user trust is improved due to the interpretability of the road selection model.

[0090] According to the method described in the above embodiments, this embodiment will further describe from the perspective of an interpretable road network selection device. The interpretable road network selection device can be specifically implemented as an independent entity, or integrated in an electronic device. The electronic device can be a terminal, a server, etc. Among them, the terminal can include a tablet computer, a notebook computer, a personal computer (PC), a microprocessing box, or other devices, etc.

[0091] Please refer to Figure 6 , Figure 6 Specifically describes the interpretable road network selection device provided by the embodiments of this application, which is applied to an electronic device. The interpretable road network selection device may include:

[0092] An acquisition module, configured to acquire the description features of each test road and construct a feature vector based on the description features;

[0093] A determination and sorting module, configured to determine a teacher model and sort the feature importance of the description features input to the teacher model through an interpretable tool;

[0094] A construction module, configured to construct an additive neural network;

[0095] A calculation module, configured to calculate a knowledge distillation loss function based on the prediction results of the teacher model;

[0096] A road selection module, configured to use the sorted description features as a training set to input into the additive neural network, train the additive neural network through the knowledge distillation loss function, use the trained additive neural network as an interpretable road network selection model, and perform road selection through the interpretable road network selection model.

[0097] In specific implementation, each of the above modules and / or units can be implemented as an independent entity, or can be arbitrarily combined and implemented as the same or several entities. For the specific implementation of each of the above modules and / or units, reference can be made to the foregoing method embodiments. For the beneficial effects that can be specifically achieved, please also refer to the beneficial effects in the foregoing method embodiments, which will not be elaborated herein.

[0098] In addition, an embodiment of the present application further provides an electronic device, which can be a device such as a computer or a tablet computer. As Figure 7 shown, the electronic device 400 includes a processor 401 and a memory 402. Among them, the processor 401 is electrically connected to the memory 402.

[0099] The processor 401 is the control center of the electronic device 400, connects various parts of the entire electronic device through various interfaces and lines, executes various functions of the electronic device and processes data by running or loading application programs stored in the memory 402 and calling data stored in the memory 402, so as to monitor the electronic device as a whole.

[0100] In this embodiment, the processor 401 in the electronic device 400 will load the instructions corresponding to the processes of one or more application programs into the memory 402 according to the following steps, and the processor 401 will run the application programs stored in the memory 402 to implement various functions:

[0101] Obtain the description features of each test road, and construct a feature vector based on the description features;

[0102] Determine the teacher model, and sort the feature importance of the description features input into the teacher model through an interpretable tool;

[0103] Construct an additive neural network;

[0104] Calculate the knowledge distillation loss function based on the prediction results of the teacher model;

[0105] Use the sorted description features as a training set and input them into the additive neural network, train the additive neural network through the knowledge distillation loss function, use the trained additive neural network as an interpretable road network selection model, and perform road selection through the interpretable road network selection model.

[0106] This electronic device can implement the steps in any embodiment of the interpretable road network selection method provided by the embodiments of the present application. Therefore, it can achieve the beneficial effects that can be achieved by any interpretable road network selection method provided by the embodiments of the present invention. For details, please refer to the foregoing embodiments, which will not be elaborated herein.

[0107] Figure 8 FIG. Figure 8 shows a specific structural block diagram of an electronic device provided by an embodiment of the present invention. The electronic device can be used to implement the interpretable road network selection method provided in the above embodiments. The electronic device 500 can be a device such as a terminal or a server. Among them, the terminal can include a tablet computer, a notebook computer, a personal computer (PC), a microprocessing box, or other devices, etc.

[0108] The RF circuit 510 is used to receive and send electromagnetic waves, realize the mutual conversion between electromagnetic waves and electrical signals, so as to communicate with a communication network or other devices. The RF circuit 510 can include various existing circuit elements for performing these functions. For example, an antenna, a radio frequency transceiver, a digital signal processor, an encryption / decryption chip, a subscriber identity module (SIM) card, a memory, and so on. The RF circuit 510 can communicate with various networks such as the Internet, an enterprise intranet, a wireless network, or communicate with other devices through a wireless network. The above-mentioned wireless network can include a cellular phone network, a wireless local area network, or a metropolitan area network. The above-mentioned wireless network can use various communication standards, protocols, and technologies, including but not limited to the Global System for Mobile Communication (GSM), Enhanced Data GSM Environment (EDGE), Wideband Code Division Multiple Access (WCDMA), Code Division Access (CDMA), Time Division Multiple Access (TDMA), Wireless Fidelity (Wi-Fi) (such as the Institute of Electrical and Electronics Engineers standards IEEE 802.11a, IEEE 802.11b, IEEE 802.11g, and / or IEEE 802.11n), Voice over Internet Protocol (VoIP), Worldwide Interoperability for Microwave Access (Wi-Max), other protocols for email, instant messaging, and short messages, and any other suitable communication protocol, and even can include those protocols that have not been developed yet.

[0109] The memory 520 can be used to store software programs and modules, such as the corresponding program instructions / modules in the above embodiments. The processor 580 executes various functional applications and data processing by running the software programs and modules stored in the memory 520, that is, realizes functions such as taking pictures with the front camera, processing the captured images, and switching the display colors of the display content on the display screen. The memory 520 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some instances, the memory 520 may further include a memory remotely disposed relative to the processor 580, and these remote memories can be connected to the electronic device 500 through a network. Examples of the above network include but are not limited to the Internet, enterprise intranet, local area network, mobile communication network, and combinations thereof.

[0110] The input unit 530 can be used to receive input digital or character information, and generate a keyboard and a mouse related to user settings and function controls.

[0111] The display unit 540 can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces, and these graphical user interfaces can be composed of graphics, text, icons, videos, and any combination thereof. The display unit 540 may include a display panel 541. Optionally, the display panel 541 can be configured in the form of an LCD (Liquid Crystal Display) or an OLED (Organic Light-Emitting Diode).

[0112] The audio circuit 560, the speaker 561, and the microphone 562 can provide an audio interface between the user and the electronic device 500. The audio circuit 560 can transmit the electrical signal converted from the received audio data to the speaker 561, and the speaker 561 converts it into a sound signal for output; on the other hand, the microphone 562 converts the collected sound signal into an electrical signal, which is received by the audio circuit 560 and then converted into audio data. After the audio data is output to the processor 580 for processing, it is sent to another terminal, for example, through the RF circuit 510, or the audio data is output to the memory 520 for further processing. The audio circuit 560 may further include an earphone jack to provide communication between the peripheral earphone and the electronic device 500.

[0113] The electronic device 500 can help the user receive requests, send information, etc. through the transmission module 570 (such as a Wi-Fi module), and it provides the user with wireless broadband Internet access. Although the transmission module 570 is shown in the figure, it can be understood that it does not belong to the essential composition of the electronic device 500 and can be omitted completely as needed without changing the essence of the invention.

[0114] The processor 580 is the control center of the electronic device 500, connecting various parts of the entire mobile phone through various interfaces and circuits. By running or executing software programs and / or modules stored in the memory 520, and by calling the data stored in the memory 520, it executes various functions of the electronic device 500 and processes data, thereby monitoring the electronic device as a whole. Optionally, the processor 580 may include one or more processing cores; in some embodiments, the processor 580 may integrate an application processor and a modem processor. Among them, the application processor mainly processes the operating system, user interface, application programs, etc., and the modem processor mainly processes wireless communications. It can be understood that the above-mentioned modem processor may not be integrated into the processor 580 either.

[0115] The electronic device 500 also includes a power supply 590 (such as a battery) for powering each component. In some embodiments, the power supply can be logically connected to the processor 580 through a power management system, so as to realize functions such as management of charging, discharging, and power consumption management through the power management system. The power supply 590 may also include any components such as one or more DC or AC power supplies, a recharge system, a power failure detection circuit, a power converter or inverter, and a power status indicator.

[0116] Although not shown, the electronic device 500 also includes a camera (such as a front camera, a rear camera), a Bluetooth module, etc., which will not be elaborated here. Specifically, in this embodiment, the display unit of the electronic device is a touch screen display, and the mobile terminal also includes a memory, and one or more programs, where one or more programs are stored in the memory and are configured to be executed by one or more processors. One or more programs include instructions for performing the following operations:

[0117] Obtain the description features of each test road, and construct a feature vector based on the description features;

[0118] Determine a teacher model, and sort the feature importance of the description features input into the teacher model through an explainable tool;

[0119] Construct an additive neural network;

[0120] Calculate a knowledge distillation loss function based on the prediction results of the teacher model;

[0121] Use the sorted description features as a training set and input them into the additive neural network. Train the additive neural network through the knowledge distillation loss function, use the trained additive neural network as an explainable road network selection model, and perform road selection through the explainable road network selection model.

[0122] In specific implementation, each of the above modules can be implemented as an independent entity, or can be arbitrarily combined and implemented as the same or several entities. For the specific implementation of each of the above modules, reference can be made to the foregoing method embodiments, which will not be elaborated herein.

[0123] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructions, or by controlling relevant hardware through instructions. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor. Therefore, an embodiment of the present invention provides a storage medium, in which multiple instructions are stored, and these instructions can be loaded by a processor to execute the steps of any one of the embodiments of the interpretable road network selection method provided by the embodiments of the present invention.

[0124] Among them, the computer-readable storage medium may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk, etc.

[0125] Since the instructions stored in this storage medium can execute the steps of any one of the embodiments of the interpretable road network selection method provided by the embodiments of the present invention, the beneficial effects achievable by any of the interpretable road network selection methods provided by the embodiments of the present invention can be realized. For details, refer to the foregoing embodiments, which will not be elaborated herein.

[0126] The above has introduced in detail an interpretable road network selection method, device, storage medium and electronic device provided by the embodiments of the present application. Specific examples are used in this article to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. An interpretable road network selection method, characterized in that: The method comprises: Obtaining descriptive features of each test road, and constructing a feature vector based on the descriptive features; the descriptive features include road descriptive features and road connectivity features, the road descriptive features at least include road grade, number of lanes, road surface material, length, width, and number of arcs, and the road connectivity features at least include degree centrality, clustering coefficient, betweenness centrality, closeness centrality, and feature vector centrality; determining a teacher model and ranking the feature importance of the descriptive features input into the teacher model by an interpretable tool; Constructing additive neural networks; The knowledge distillation loss function is calculated based on the prediction result of the teacher model; the calculation of the knowledge distillation loss function based on the prediction result of the teacher model includes: predicting the first probability of each test road being selected and retained by the teacher model, and marking the first probability as a first label; calculating the soft label of the test road selected and retained required for knowledge distillation based on the first label; calculating the knowledge distillation loss function based on the soft label; wherein the soft label is calculated by a first formula, and the first formula is: in, is the first label, is the distillation temperature; The knowledge distillation loss function is calculated by the second formula, and the second formula is: in, is the prediction result of the additive neural network, To test the manual labeling of roads, is the weight parameter of soft label loss and true label loss, is the mean square error function; The sorted description features are input into the additive neural network as a training set, the additive neural network is trained through the knowledge distillation loss function, the trained additive neural network is used as an interpretable road network selection model, and road selection is performed through the interpretable road network selection model.

2. The interpretable road network selection method according to claim 1, characterized in that: The determining of the teacher model comprises: The feature vector is used as a training set and a test set, a plurality of different machine learning models are trained through the training set, and the test set is input into the trained plurality of different machine learning models to obtain prediction results, and the machine learning model with the highest prediction result accuracy is selected as the teacher model.

3. The interpretable road network selection method according to claim 1, characterized in that: The step of ranking the feature importance of the descriptive features input into the teacher model by an interpretable tool comprises: Calculate the j The average contribution of a feature to all test samples: in, represents the total number of test roads, Indicates i The first target test road j SHAP contribution value of each feature; The description features are sorted in descending order according to the SHAP contribution values.

4. The interpretable road network selection method according to claim 1, characterized in that: The additive neural network includes an input layer, an intermediate layer and an output layer, and the output of the additive neural network is used to determine whether the test road is selected and retained; The mathematical expression of the additive neural network is: in, represents the feature neural network, n Represents the number of feature networks, and its maximum value is the total number of described features.

5. An interpretable road network selection device, the interpretable road network selection device is used to implement the interpretable road network selection method according to claim 1, characterized in that: include: An acquisition module, used for acquiring descriptive features of each test road and constructing a feature vector based on the descriptive features; A determination and ranking module, for determining a teacher model and ranking the feature importance of the descriptive features input into the teacher model through an interpretable tool; Building blocks for network construction of additive neural networks; A calculation module, used for calculating a knowledge distillation loss function based on the prediction result of the teacher model; A road selection module is used to input the sorted description features as a training set into the additive neural network, train the additive neural network through the knowledge distillation loss function, use the trained additive neural network as an interpretable road network selection model, and perform road selection through the interpretable road network selection model.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the interpretable road network selection method according to any one of claims 1 to 4.

7. An electronic device, characterized in that: It includes a processor and a memory, the processor is electrically connected to the memory, the memory is used to store instructions and data, and the processor is used to execute the steps in the interpretable road network selection method described in any one of claims 1 to 4.

Images