Pet feeding plan processing method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a pet feeding plan processing method and device, electronic equipment and a storage medium. The method comprises the following steps: receiving a pet feeding plan generation instruction of a first pet, and obtaining first pet attribute information and first pet behavior sequence information of the first pet according to the pet feeding plan generation instruction; determining a pet community where the first pet is located according to the first pet attribute information; and calling a pet feeding prediction model corresponding to the pet community based on the first pet attribute information and the first pet behavior sequence information to perform pet feeding prediction, and generating a pet feeding plan of the first pet according to a prediction result. The method can improve the accuracy of pet feeding plan processing.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to a pet feeding plan processing method, device, electronic device, and storage medium. Background Art

[0002] In recent years, with the continuous development of science and technology and the accelerating pace of life, people's life pressure has been continuously increasing. The companionship of pets can greatly relieve people's mental stress and bring them physical and mental pleasure, which has led to the rapid development of the pet products market in recent years.

[0003] Among them, the pet feeding equipment in pet supplies can automatically add food and water to the pet by setting a pet feeding plan, thereby reducing the pet owner's pressure on raising pets and improving the happiness of raising pets.

[0004] However, in the related art, the accuracy of formulating a feeding plan for a pet is not high. Summary of the Invention

[0005] The main purpose of the embodiments of the present application is to propose a pet feeding plan processing method, device, electronic device and storage medium, aiming to improve the accuracy of pet feeding plan processing.

[0006] To achieve the above objectives, a first aspect of an embodiment of the present application provides a method for processing a pet feeding plan, the method comprising: receiving a pet feeding plan generation instruction for a first pet, and acquiring first pet attribute information and first pet behavior sequence information of the first pet according to the pet feeding plan generation instruction; determining the pet community where the first pet is located according to the first pet attribute information; Based on the first pet attribute information and the first pet behavior sequence information, a pet eating prediction model corresponding to the pet community is called to perform pet eating prediction, and a pet feeding plan for the first pet is generated according to the prediction result. The pet eating prediction model is trained based on pet eating statistical data of multiple second pets in the pet community.

[0007] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a pet feeding plan processing device, the device comprising: a receiving unit, configured to receive a pet feeding plan generation instruction for a first pet, and acquire first pet attribute information and first pet behavior sequence information of the first pet according to the pet feeding plan generation instruction; a determining unit, configured to determine the pet community where the first pet is located according to the first pet attribute information; A generation unit is used to call a pet eating prediction model corresponding to the pet community to perform pet eating prediction based on the first pet attribute information and the first pet behavior sequence information, and generate a pet feeding plan for the first pet according to the prediction result, wherein the pet eating prediction model is trained based on pet eating statistical data of multiple second pets in the pet community.

[0008] In some embodiments, the present application provides a model training device, comprising: a first acquiring subunit, configured to acquire second pet attribute information, second pet behavior sequence information, and pet eating statistics of each of the second pets in the pet community, wherein the pet eating statistics and the second pet behavior sequence information are collected in the same time period; The first adjustment subunit is used to adjust the parameters of the pre-trained neural network model using the second pet attribute information of the multiple second pets and the second pet behavior sequence information as sample input, and using the pet eating statistics of the corresponding second pets as sample labels to obtain a pet eating prediction model.

[0009] Optionally, in some embodiments, the first adjustment subunit includes: A prediction module, configured to call a pre-trained neural network model to perform pet eating prediction based on second pet attribute information of a plurality of second pets and corresponding second pet behavior sequence information, and obtain a plurality of pet eating prediction data corresponding to the plurality of second pets; a construction module, configured to construct a loss function based on the plurality of pet eating prediction data and the pet eating statistical data of the plurality of second pets; An updating module is used to perform multiple rounds of iterative updates on the model parameters of the neural network model according to the loss function, and to determine a pet eating prediction model based on the model parameters when a preset convergence condition is reached.

[0010] Optionally, in some embodiments, the present application further provides a model training device, comprising: a second acquiring subunit, configured to acquire second pet attribute information, second pet behavior sequence information, first weather information, and pet eating statistics of each of the second pets in the pet community, wherein the pet eating statistics, the first weather information, and the second pet behavior sequence information are collected in the same time period; a second adjustment subunit, configured to use the second pet attribute information of the plurality of second pets, the first weather information, and the second pet behavior sequence information as sample inputs, and use the pet eating statistics of the corresponding second pets as sample labels to adjust parameters of the pre-trained neural network model to obtain a pet eating prediction model; The generating unit is further configured to: The second weather information is obtained, and the pet eating prediction model is called to perform pet eating prediction based on the second weather information, the first pet attribute information and the first pet behavior sequence information, and a pet feeding plan for the first pet is generated according to the prediction result.

[0011] Optionally, in some embodiments, the receiving unit includes: a generating subunit, configured to receive a pet feeding plan generating instruction for a first pet, obtain first time period information of a feeding plan application from the pet feeding plan generating instruction, and generate pet attribute information of the first pet based on the first time period information; The third acquiring subunit is configured to acquire first pet behavior sequence information of the first pet within a time range corresponding to a preset second time period information.

[0012] Optionally, in some embodiments, the receiving subunit includes: A first display module is used to display an application interface of a pet feeding device management application; a second display module, configured to display a pet feeding plan creation interface for the first pet in response to a trigger operation for creating a feeding plan for the first pet received in the application interface, the pet feeding plan creation interface including an automatic creation control; A generating module is configured to generate a pet feeding plan generating instruction for a first pet in response to a triggering operation on the automatic creation control.

[0013] Optionally, the pet feeding plan processing device provided by the present application further includes: A third display module is used to display a pet feeding plan generation prompt on the application interface; a fourth display module, configured to display a display page of the pet feeding plan in response to a triggering operation for generating a prompt for the pet feeding plan, the display page including a feeding plan application control; An application module is configured to apply the pet feeding plan to the feeding device in response to a triggering operation on the feeding plan application control.

[0014] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the pet feeding plan processing method described in the first aspect.

[0015] To achieve the above-mentioned purpose, the fourth aspect of the embodiments of the present application proposes a storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the pet feeding plan processing method described in the first aspect.

[0016] To achieve the above-mentioned purpose, the fifth aspect of the embodiment of the present application proposes a computer program product, which includes a computer program, and the computer program is read and executed by the processor of a computer device, so that the computer device executes the pet feeding plan processing method described in the first aspect.

[0017] The pet feeding plan processing method proposed in the embodiment of the present application receives a pet feeding plan generation instruction of a first pet, and obtains first pet attribute information and first pet behavior sequence information of the first pet according to the pet feeding plan generation instruction; determines the pet community in which the first pet is located according to the first pet attribute information; calls a pet eating prediction model corresponding to the pet community based on the first pet attribute information and the first pet behavior sequence information to perform pet eating prediction, and generates a pet feeding plan for the first pet according to the prediction result, wherein the pet eating prediction model is trained based on pet eating statistical data of multiple second pets in the pet community.

[0018] As can be seen from this, the pet feeding plan processing method provided in the embodiments of the present application can first obtain the eating statistics of community pets based on the pet community where the pet lives, then train a pet eating prediction model based on the eating statistics of community pets, and then generate a corresponding feeding plan for the pet based on the pet eating model, thereby obtaining an accurate pet feeding plan. This method can improve the accuracy of pet feeding plan generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the technical solution of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present application and do not constitute a limitation on the technical solution of the present application.

[0020] Figure 1 A schematic diagram of a system used in the pet feeding plan processing method provided in this application; Figure 2 A flowchart of the pet feeding plan processing method provided for this application; Figure 3 A schematic diagram of an interface for a pet feeding device management application provided in this application; Figure 4 A schematic diagram of the feeding plan creation interface in the pet feeding device management application provided in this application; Figure 5Another flowchart of the pet feeding plan processing method provided in this application; Figure 6 Another flowchart of the pet feeding plan processing method provided in this application; Figure 7 A schematic diagram of the structure of the pet feeding plan processing device provided in this application; Figure 8 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0022] Before further explaining the embodiments of the present application in detail, the nouns and terms involved in the embodiments of the present application are explained. The nouns and terms involved in the embodiments of the present application are subject to the following interpretations: Neural network models: Since the advent of mathematical methods that simulate actual human neural networks, people have gradually become accustomed to simply referring to these artificial neural networks as neural networks. Neural networks hold broad and attractive prospects in areas such as system identification, pattern recognition, and intelligent control. In intelligent control, people are particularly interested in the self-learning capabilities of neural networks, and view this important feature of neural networks as one of the keys to solving the difficult problem of controller adaptability in automatic control.

[0023] Large Language Models (LLMs) are deep learning models trained using large amounts of text data, enabling them to generate natural language text or understand the meaning of text. These models can be trained on massive datasets to provide in-depth knowledge and language production across a wide range of topics. Their core concept is to learn the patterns and structure of natural language through large-scale unsupervised training, simulating, to a certain extent, the human language cognition and generation process.

[0024] In the related art, when pet owners set feeding plans for automatic pet feeding devices, they generally do so manually based on experience, such as setting fixed feeding times and feeding amounts for each feeding, and setting feeding plans to cycle weekly. However, this feeding plan is completely manually generated, and in some cases, pet owners do not understand their pets' actual appetite, which can lead to inaccurate feeding plan settings. If the amount of food fed is too much, the food may not be consumed in time, resulting in long-term exposure to air and becoming stale, affecting the pet's appetite and even affecting the pet's health.

[0025] Based on this, and to address a series of issues caused by inaccurate pet feeding plan settings in related technologies, the present invention provides a pet feeding plan processing method, device, electronic device, and storage medium. This method uses artificial intelligence technology combined with the eating habits of pets in the pet community to generate an accurate pet feeding plan. The following describes the pet feeding plan processing method provided in the present invention.

[0026] Reference Figure 1 , which is a schematic diagram of a system used in the pet feeding plan processing method provided herein. As shown, the system includes at least one pet feeding device 110, at least one pet device management client 130, and a pet device management server 120. The pet feeding device 110 can be a pet food feeding management device or a pet water feeding management device. The pet device management client 130 is specifically a terminal loaded with a pet device management application. This terminal can be a mobile terminal such as a smartphone, or various other terminals such as a tablet, personal computer, head-mounted device, or vehicle-mounted terminal. In addition to managing pet feeding devices, the pet device management application can also control other pet devices such as pet excretion management devices and pet toys. The pet device management server 120 can be a server that provides management, data processing, and interaction services to multiple pet device management clients 130. The pet device management server 120 can be a single high-performance computer in a network platform, a cluster of multiple high-performance computers, a portion of a single high-performance computer (e.g., a virtual machine), or a combination of portions of multiple high-performance computers (e.g., virtual machines). The pet device management server 120 can be a physical server or a cloud server.

[0027] In an embodiment of the present application, a pet owner can initiate a pet feeding plan generation instruction via the pet device management client 130. Upon receiving this instruction, the pet device management server 120 can retrieve from the instruction the location information and pet attribute information of the pet for which a feeding plan is to be generated, as well as information about the pet's behavior sequences from various pet devices (including but not limited to the aforementioned pet feeding devices, pet excrement management devices, and pet toys). Furthermore, the server 120 can retrieve the pet's eating statistics and other related information from the pet devices corresponding to other pets in the same pet community, train a pet eating prediction model, and then use the pet eating prediction model to predict the pet's food intake and generate a pet feeding plan. Subsequently, under the instruction of the pet device management client 130, the pet device management server 120 sends the pet feeding plan to the corresponding pet feeding device 110 for application.

[0028] Reference Figure 2 , which is a flow chart of the pet feeding plan processing method provided in this application.

[0029] In some embodiments, the pet feeding plan processing method provided in the embodiments of the present application includes but is not limited to steps S201 to S204.

[0030] Step S201, receiving a pet feeding plan generation instruction for a first pet, and acquiring first pet attribute information and first pet behavior sequence information according to the pet feeding plan generation instruction; Step S202, determining the pet community where the first pet is located based on the first pet attribute information; Step S203 : Based on the first pet attribute information and the first pet behavior sequence information, a pet eating prediction model corresponding to the pet community is called to perform pet eating prediction, and a pet feeding plan for the first pet is generated according to the prediction result.

[0031] In steps S201 to S203 shown in the embodiment of the present application, upon receiving the pet feeding plan generation instruction, the pet device management server can extract the corresponding pet information from the pet feeding plan generation instruction. To distinguish it from other pets, the pet targeted by the feeding plan is referred to as the first pet. Then, the attribute information of the pet and the behavior sequence information of the pet can be obtained. Then, the pet community to which the first pet is located is determined based on the first attribute information of the first pet. Specifically, the pet community to which the first pet belongs can be determined based on the species and breed of the first pet. Pet communities can be divided according to biological categories, for example, into cat communities, dog communities, etc., or can be further subdivided into blue cat communities, orange cat communities, etc. Pets in different pet communities are relatively similar in terms of physiological characteristics, eating habits, etc., so that a more accurate feeding prediction model can be trained.

[0032] In this way, after determining the pet community in which the first pet is located, a pet eating prediction model corresponding to the pet community can be determined based on the pet community. The pet eating prediction model can then be called based on the pet attribute information and pet behavior sequence information of the first pet to predict the pet's eating habits, obtaining a feeding prediction result for the first pet. An accurate pet feeding plan for the first pet can then be generated based on the feeding prediction result. The pet behavior sequence information reveals information such as the pet's behavioral habits and exercise volume over the past period of time. Based on the pet's real-time exercise volume, the pet's energy consumption and the amount of food and water required can be accurately assessed. Furthermore, the pet feeding device 110 in the embodiment of the present application can be equipped with multiple image acquisition devices for real-time behavioral image acquisition of the first pet, and then the pet behavior sequence is determined based on the real-time acquired pet behavior image sequence.

[0033] Among them, for different pet communities, their corresponding pet eating prediction models can be trained independently. Therefore, after determining the pet community where the first pet is located, the pet eating prediction model corresponding to the pet community can be selected based on the pet community. This can make the eating prediction model more targeted, thereby improving the accuracy of the eating prediction. Among them, the process of training the pet eating model can be to obtain pet eating statistics based on multiple second pets in the pet community, and then use the pet eating statistics of the second pets in the pet community to train the pet eating prediction model. Compared with the pet eating prediction model trained with a large amount of pet eating statistics without community restrictions, it can make more accurate pet food intake predictions for pets in this pet community.

[0034] In this way, the pet feeding plan processing method provided by this application can classify the pet community in advance, and then call the eating prediction model trained based on the eating statistics of other pets in the community when generating the pet feeding plan. The eating prediction model is used to generate the pet feeding plan based on the pet's behavior sequence, thereby greatly improving the accuracy of the pet feeding plan formulation.

[0035] In some embodiments, receiving a pet feeding plan generation instruction for a first pet, and obtaining first pet attribute information and first pet behavior sequence information of the first pet according to the pet feeding plan generation instruction, includes: receiving a pet feeding plan generation instruction for a first pet, obtaining first time period information for a feeding plan application from the pet feeding plan generation instruction, and generating pet attribute information for the first pet based on the first time period information; First pet behavior sequence information of the first pet within a time range corresponding to the preset second time period information is obtained.

[0036] Embodiments of the present application can generate a pet feeding plan for the current moment or a short period of time after the current moment (e.g., a week or a month), or can generate a pet feeding plan that may be needed in the future (e.g., a feeding plan for three months or a year from now). Therefore, when obtaining first pet attribute information for a first pet, the time period information corresponding to the feeding plan contained in the feeding plan generation instruction can be first obtained. To distinguish it from other time period information in this application, this information may be referred to as first time period information. When the first time period information indicates a short period of time after the current moment, the current attribute information of the first pet can be directly obtained. The first pet attribute information of the first pet may include, but is not limited to, information related to the pet's own attributes, such as the pet's species (e.g., cat or dog), breed (e.g., golden retriever, corgi, border collie), gender, age, weight, and historical medical records. If the first time period information indicates a period of time in the future, the pet attribute information can be divided into variable attributes and immutable attributes. Immutable attributes may include attributes such as the pet's species, breed, and gender, while variable attributes may include attributes such as the pet's age and weight. Then, the variable attributes may be predicted, and then first pet attribute information of the first pet may be determined based on the predicted attributes and the immutable attributes.

[0037] In this embodiment of the present application, a pet's diet is not only related to the pet's attributes but also to its behavioral habits. For example, a pet who loves exercise generally eats more than a pet who doesn't, and exercise levels vary between pets who love exercise. Therefore, this embodiment not only acquires the attribute information of a first pet, but also acquires the first pet's behavioral sequence information. Because a pet's exercise habits may vary with factors such as age and physical condition, it is necessary to acquire the pet's behavioral sequence information for a second time period that is close to the first time period, such as the first pet's behavioral sequence information within a week or month prior to the current time. In this embodiment of the present application, the first pet's behavioral sequence information can be captured using a camera module provided in a pet feeding device. In some embodiments, a pet-owning household may have multiple pet devices. These multiple pet devices can be of the same type, such as multiple pet feeding devices, or different types, such as pet feeding devices, pet waste disposal devices, and pet toys. Each of these multiple pet devices is equipped with a camera module, allowing the pet's behavioral sequence to be captured from multiple angles on multiple devices, thereby obtaining richer pet behavior sequence information.

[0038] In some embodiments, the process of generating instructions for generating a pet feeding plan for a first pet includes: Display the application interface of the pet feeding equipment management application; In response to a trigger operation received in the application interface to create a feeding plan for the first pet, displaying a pet feeding plan creation interface for the first pet, the pet feeding plan creation interface including an automatic creation control; In response to a triggering operation on the automatic creation control, a pet feeding plan generation instruction for the first pet is generated.

[0039] In an embodiment of the present application, the pet feeding plan generation instruction can be initiated by the user in the application interface of the pet feeding device management application. Figure 3 , which is a schematic diagram of an interface for the pet feeding device management application provided herein. As shown, when a user opens the pet feeding device management application, an application interface 310 may be displayed, within which a feeding plan creation control 311 may be displayed. The user can trigger a feeding plan creation interface for a first pet by triggering feeding plan creation control 311.

[0040] See also Figure 4 , which is a schematic diagram of the feeding plan creation interface in the pet feeding device management application provided by this application. As shown in the figure, in the pet feeding plan creation interface 410, an edit plan control and an automatic creation control 411 can be displayed. When the user clicks the automatic creation control 411, the generation of a pet feeding plan generation instruction for the first pet can be triggered. In some embodiments, when the pet feeding device is bound to multiple pets, after clicking the automatic creation control 411, the pet selection interface can be entered, and after selecting the first pet in this interface, the generation of a pet feeding plan generation instruction for the first pet is triggered. Alternatively, when it is necessary to generate a pet feeding plan for a future time period, after clicking the automatic creation control 411, the time period setting interface can be displayed first. After receiving the first time period information entered in the time period setting interface, the generation of a pet feeding plan generation instruction for the first pet is triggered according to the first time period information.

[0041] See also Figure 5 , is another flow chart of the pet feeding plan processing method provided by the present application. In the embodiment of the present application, the pet feeding plan processing method provided by the embodiment of the present application includes but is not limited to the steps of: Step S201, receiving a pet feeding plan generation instruction for a first pet, and acquiring first pet attribute information and first pet behavior sequence information of the first pet according to the pet feeding plan generation instruction; Step S202, determining the pet community where the first pet is located based on the first pet attribute information; Step S2031, obtaining second pet attribute information, second pet behavior sequence information, and pet eating statistics of each second pet in the pet community, where the pet eating statistics and the second pet behavior sequence information are collected in the same time period; Step S2032, using second pet attribute information and second pet behavior sequence information of multiple second pets as sample inputs, and using corresponding pet eating statistics of the second pets as sample labels, adjusting parameters of the pre-trained neural network model to obtain a pet eating prediction model; Step S204: calling a pet eating prediction model based on the first pet attribute information and the first pet behavior sequence information to perform pet eating prediction, and generating a pet feeding plan for the first pet according to the prediction result.

[0042] In the embodiment of the present application, the detailed process of training the pet eating prediction model based on the pet eating statistics of multiple second pets is introduced. Among them, step S201, step S202 and step S204 have been introduced above and will not be repeated here.

[0043] Specifically, after determining the pet community in which the first pet resides, information about multiple second pets associated with the pet community can be obtained based on the pet community information. These second pets can be pets in the pet community. Then, the second pet attribute information, second pet behavior sequence information, and pet eating statistics for each second pet are obtained. The method for obtaining the second pet attribute information and the second pet behavior sequence information can be consistent with the method for obtaining the first pet attribute information and the first pet behavior sequence information described above, and will not be further described here. The eating statistics for the second pets can specifically be eating statistics collected during the same time period as the second pet behavior sequence information. For example, the same time period can be a single day, multiple one-day periods, or a period of multiple consecutive days. The collection of the eating statistics for the second pets can specifically be performed using a feeding device corresponding to each second pet. In addition to being equipped with a camera module, the feeding device provided in this application can also include a weighing module and a pet behavior analysis module. The pet behavior analysis module can analyze whether the pet's behavior is eating based on photos or videos of the pet's behavior captured by the camera module. When the pet's eating behavior is identified, the module records the pet's eating time and the start and end times of the pet's eating behavior. The module then calculates the difference in weight of the pet's food bowl at these two time points to determine the pet's food intake. Alternatively, in some embodiments, the feeding device can first collect preliminary eating statistics, which can then be corrected by the pet owner based on the eating video corresponding to the eating statistics to obtain accurate eating statistics.

[0044] After obtaining the second pet attribute information, second pet behavior sequence information, and pet eating statistics corresponding to each second pet of multiple second pets, the second pet attribute information and second pet behavior sequence information of the second pets can be used as model input, and the pet eating statistics corresponding to each second pet can be used as model output to train the pet eating prediction model. The model architecture of the pet eating prediction model can be a multi-task learning model, which includes a pet eating time prediction branch and a pet eating amount prediction branch. When training the model, the labels corresponding to these two branches can be extracted from the pet eating statistics, and then the parameters of the model can be adjusted based on the labels of the two branches. The multi-task learning model in the present application can be a pre-trained model. Specifically, a large amount of pet basic attribute data and pet behavior sequence data can be used to perform unsupervised training on the model, thereby improving the feature extraction capability of the model.

[0045] In some embodiments, during the training phase, pet eating prediction models corresponding to different pet communities can utilize a variety of machine learning and deep learning models (including but not limited to random forests, support vector machines (SVMs), XGBoost, long short-term memory networks (LSTMs), convolutional neural networks (CNNs), etc.). For each pet community, multiple models can be trained separately. The trained pet eating prediction models are then comprehensively evaluated using metrics such as cross-validation, mean squared error (MSE), root mean squared error (RMSE), and accuracy. Ultimately, the model with the best prediction performance is selected as the pet eating prediction model for that community.

[0046] In some embodiments, the parameters of a pre-trained neural network model are adjusted using second pet attribute information and second pet behavior sequence information of multiple second pets as sample inputs, and pet eating statistics of corresponding second pets as sample labels to obtain a pet eating prediction model, including: Based on the second pet attribute information of the plurality of second pets and the corresponding second pet behavior sequence information, a pre-trained neural network model is called to predict the pet's eating habits, and a plurality of pet eating prediction data corresponding to the plurality of second pets is obtained; Constructing a loss function based on the plurality of pet eating prediction data and the pet eating statistics of the plurality of second pets; The model parameters of the neural network model are iteratively updated for multiple rounds according to the loss function, and the pet eating prediction model is determined based on the model parameters when the preset convergence conditions are reached.

[0047] In an embodiment of the present application, the process of supervised fine-tuning of the pre-trained multi-task learning model can specifically include inputting the second pet attribute information of multiple second pets and the corresponding second pet behavior sequence information into the pre-trained neural network model to perform pet eating prediction, thereby obtaining multiple pet eating prediction data corresponding to the multiple second pets. Wherein, when the neural network model is a multi-task learning model, each second pet corresponds to eating prediction data containing multiple task branches. Furthermore, a loss function can be constructed and the loss value can be calculated using the cross entropy method based on the eating prediction data and the corresponding label data, and then the back propagation gradient can be calculated based on the loss value, and the model parameters of the multi-task learning model can be iteratively updated according to the back propagation gradient until the model converges, thereby obtaining a trained pet eating prediction model. Wherein, the model convergence here can specifically be that the model training reaches a preset convergence condition, such as the parameter change amplitude of the model is less than the preset amplitude, or the iterative update rounds of the model parameters have reached the preset rounds.

[0048] See also Figure 6 , is another flow chart of the pet feeding plan processing method provided by the present application. In the embodiment of the present application, the pet feeding plan processing method provided by the embodiment of the present application includes but is not limited to the steps of: Step S201, receiving a pet feeding plan generation instruction for a first pet, and acquiring first pet attribute information and first pet behavior sequence information of the first pet according to the pet feeding plan generation instruction; Step S202, determining the pet community where the first pet is located based on the first pet attribute information; Step S2033, obtaining second pet attribute information, second pet behavior sequence information, first weather information, and pet eating statistics of each of the second pets in the pet community, wherein the pet eating statistics, the first weather information, and the second pet behavior sequence information are collected in the same time period; Step S2034, using the second pet attribute information of the plurality of second pets, the first weather information, and the second pet behavior sequence information as sample inputs, and using the corresponding pet eating statistics of the second pets as sample labels, adjusting the parameters of the pre-trained neural network model to obtain a pet eating prediction model; Step S205, obtaining the second weather information, and calling the pet eating prediction model to perform pet eating prediction based on the second weather information, the first pet attribute information and the first pet behavior sequence information, and generating a pet feeding plan for the first pet according to the prediction result.

[0049] Environmental differences among pets can include climate (including but not limited to weather, temperature, and humidity), air quality (such as greenery coverage and air quality index), functional divisions (such as industrial parks, residential areas, and commercial districts), and distribution of activity areas (such as the number of parks, especially pet parks). Even pets of the same type can eat differently in different living environments. However, pets of the same type in the same or similar environments can maintain a certain degree of similarity in their food intake. Therefore, when training a pet food intake prediction model, factors such as environmental and geographic location can be used as training features to improve the accuracy of the model.

[0050] In an embodiment of the present application, when training a pet eating prediction model, weather information can be added to the model input data. Since weather information can affect factors such as a pet's mood and appetite, causing changes in the pet's food intake, adding weather information to the input data allows the pet eating prediction model to predict more accurate pet eating data, thereby enabling the formulation of a more accurate pet feeding plan. In some embodiments, after obtaining second pet attribute information, second pet behavior sequence information, and pet eating statistics corresponding to multiple second pets, geographic location information and climate information for each second pet can be further obtained. The second pet attribute information, second pet behavior sequence information, geographic location information, and climate information of the second pets can then be used as model input to train the pet eating prediction model.

[0051] Furthermore, in some embodiments, the pet eating prediction model can be trained using an incremental learning strategy, meaning it can be regularly optimized and updated with new data to adapt to changes in pet eating habits. Specifically, each pet community model is retrained or fine-tuned using the latest pet eating statistics at regular intervals (e.g., monthly or quarterly) to improve prediction accuracy.

[0052] In some embodiments, after calling a pet eating prediction model based on the first pet attribute information and the first pet behavior sequence information to predict the pet eating, and generating a pet feeding plan for the first pet according to the prediction result, the method further includes: Display pet feeding plan generation prompts on the app interface; In response to a triggering operation for generating a prompt for a pet feeding plan, displaying a display page for the pet feeding plan, the display page including a feeding plan application control; In response to a triggering operation of the feeding plan application control, the pet feeding plan is applied to the feeding device.

[0053] In an embodiment of the present application, after the pet device management server 120 generates a pet feeding plan, it can further send the generated pet feeding plan to the corresponding pet device management client 130. After receiving the generated pet feeding plan, the pet device management client 130 can display a pet feeding plan generation prompt in the application interface, indicating that the pet feeding plan has been automatically generated. The user can then click on the pet feeding plan generation prompt to further display a pet feeding plan display page. This display page displays detailed information about the generated pet feeding plan, including but not limited to data such as the feeding start time and the feeding amount. The display page can also display a feeding plan application control, which the user can click to apply the generated pet feeding plan to the pet feeding device.

[0054] In some embodiments, a modification control may also be displayed on the display page of the pet feeding plan, and the user may review the generated pet feeding plan. When the displayed pet feeding plan needs to be modified, the user may click the modification control to enter the pet feeding plan modification interface, and then modify the pet feeding plan in the modification interface.

[0055] In summary, the pet feeding plan processing method proposed in the embodiment of the present application receives a pet feeding plan generation instruction of a first pet, and obtains the first pet attribute information and the first pet behavior sequence information according to the pet feeding plan generation instruction; determines the pet community in which the first pet is located according to the first pet attribute information; obtains pet eating statistics of multiple second pets in the pet community, and trains a pet eating prediction model based on the pet eating statistics; calls the pet eating prediction model based on the first pet attribute information and the first pet behavior sequence information to perform pet eating prediction, and generates a pet feeding plan for the first pet according to the prediction result.

[0056] It can be seen from this that the pet feeding plan processing method provided in the embodiment of the present application can first obtain the eating statistics of community pets based on the pet community where the pet is located when generating a pet feeding plan, and then train the pet eating prediction model based on the eating statistics of community pets, and then generate a corresponding feeding plan for the pet based on the pet eating model, so as to obtain an accurate pet feeding plan.

[0057] Reference Figure 7 In some embodiments, the present application also provides a pet feeding plan processing device 700, which includes: The receiving unit 710 is configured to receive a pet feeding plan generation instruction for a first pet, and obtain the first pet's location information, first pet attribute information, and first pet behavior sequence information according to the pet feeding plan generation instruction; A determining unit 720 is configured to determine the pet community where the first pet is located based on the location information; The generation unit 730 is used to call the pet eating prediction model corresponding to the pet community based on the first pet attribute information and the first pet behavior sequence information to perform pet eating prediction, and generate a pet feeding plan for the first pet according to the prediction result. The pet eating prediction model is trained based on the pet eating statistical data of multiple second pets in the pet community.

[0058] In some embodiments, the present application further provides a model training device, comprising: A first acquisition subunit is configured to acquire second pet attribute information, second pet behavior sequence information, and pet eating statistics of each second pet in the pet community, where the pet eating statistics and the second pet behavior sequence information are collected in the same time period; The first adjustment subunit is used to adjust the parameters of the pre-trained neural network model using the second pet attribute information and second pet behavior sequence information of multiple second pets as sample input, and using the pet eating statistics of the corresponding second pets as sample labels to obtain a pet eating prediction model.

[0059] Optionally, in some embodiments, the first adjustment subunit includes: A prediction module, configured to call a pre-trained neural network model to perform pet eating prediction based on second pet attribute information of the plurality of second pets and corresponding second pet behavior sequence information, and obtain a plurality of pet eating prediction data corresponding to the plurality of second pets; A construction module, configured to construct a loss function based on a plurality of pet eating prediction data and a plurality of pet eating statistics of a second pet; The updating module is used to perform multiple rounds of iterative updates on the model parameters of the neural network model according to the loss function, and determine the pet eating prediction model based on the model parameters when the preset convergence conditions are reached.

[0060] Optionally, in some embodiments, the present application further provides a model training device, comprising: A second acquisition subunit is configured to acquire second pet attribute information, second pet behavior sequence information, first weather information, and pet eating statistics of each second pet in the pet community, wherein the pet eating statistics, first weather information, and second pet behavior sequence information are collected in the same time period; a second adjustment subunit, configured to use second pet attribute information of a plurality of second pets, first weather information, and second pet behavior sequence information as sample inputs, and use pet eating statistics of corresponding second pets as sample labels to adjust parameters of a pre-trained neural network model to obtain a pet eating prediction model; Generates units, also used for: The second weather information is obtained, and a pet eating prediction model is called to perform pet eating prediction based on the second weather information, the first pet attribute information and the first pet behavior sequence information, and a pet feeding plan for the first pet is generated according to the prediction result.

[0061] Optionally, in some embodiments, the receiving unit includes: a receiving subunit, configured to receive a pet feeding plan generation instruction for a first pet, determine a feeding device corresponding to the first pet according to the pet feeding plan generation instruction, and determine location information of the first pet according to the positioning of the feeding device; a generating subunit, configured to obtain first time period information of a feeding plan application from the pet feeding plan generating instruction, and generate pet attribute information of the first pet based on the first time period information; The third acquiring subunit is configured to acquire first pet behavior sequence information of the first pet within a time range corresponding to the preset second time period information.

[0062] Optionally, in some embodiments, the receiving subunit includes: A first display module is used to display an application interface of a pet feeding device management application; a second display module, configured to display a pet feeding plan creation interface for the first pet in response to a trigger operation for creating a feeding plan for the first pet received in the application interface, the pet feeding plan creation interface including an automatic creation control; A generating module is configured to generate a pet feeding plan generating instruction for a first pet in response to a triggering operation on the automatic creation control.

[0063] Optionally, the pet feeding plan processing device provided by the present application further includes: The third display module is used to display the pet feeding plan generation prompt on the application interface; a fourth display module, configured to display a display page of the pet feeding plan in response to a triggering operation for generating a prompt for a pet feeding plan, the display page including a feeding plan application control; The application module is configured to apply the pet feeding plan to the feeding device in response to a triggering operation on the feeding plan application control.

[0064] Reference Figure 8 , Figure 8 The hardware structure of an electronic device according to another embodiment is shown. The electronic device includes: The processor 801 may be implemented as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application. The memory 802 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 802 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 802 and is called by the processor 801 to execute the pet feeding plan processing method of the embodiments of this application. Input / output interface 803, used to implement information input and output; Communication interface 804, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.); Bus 805 , which transmits information between various components of the device (e.g., processor 801 , memory 802 , input / output interface 803 , and communication interface 804 ); The processor 801 , the memory 802 , the input / output interface 803 and the communication interface 804 are connected to each other in communication within the device via a bus 805 .

[0065] An embodiment of the present application further provides a storage medium storing a computer program, which, when executed by a processor, implements the pet feeding plan processing method provided in the present application.

[0066] The present application also provides a computer program product, which includes a computer program. A processor of a computer device reads and executes the computer program, so that the computer device executes the above-mentioned pet feeding plan processing method.

[0067] The terms "first," "second," "third," "fourth," and the like (if any) in the specification of the present disclosure and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of the present disclosure described herein, for example, can be implemented in orders other than those illustrated or described herein. In addition, the terms "comprises" and "comprising," and any variations thereof, are intended to cover non-exclusive inclusions, e.g., a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or apparatus.

[0068] It should be understood that in the present disclosure, "at least one (item)" refers to one or more, and "plurality" refers to two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0069] It should be understood that in the description of the embodiments of the present application, the meaning of multiple (or multiple items) is more than two, greater than, less than, exceed, etc. are understood to exclude the number itself, and above, below, within, etc. are understood to include the number itself.

[0070] In the several embodiments provided in the present disclosure, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0071] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0072] In addition, the functional units in the various embodiments of the present disclosure may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0073] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present disclosure is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the various embodiments of the present disclosure. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0074] It should also be understood that the various implementation methods provided in the embodiments of the present application can be combined arbitrarily to achieve different technical effects.

[0075] The above is a specific description of the implementation methods of the present disclosure, but the present disclosure is not limited to the above implementation methods. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present disclosure. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present disclosure.

Claims

1. A pet feeding plan processing method, characterized in that: The method comprises: receiving a pet feeding plan generation instruction for a first pet, and acquiring first pet attribute information and first pet behavior sequence information of the first pet according to the pet feeding plan generation instruction; determining the pet community where the first pet is located according to the first pet attribute information; Based on the first pet attribute information and the first pet behavior sequence information, a pet eating prediction model corresponding to the pet community is called to perform pet eating prediction, and a pet feeding plan for the first pet is generated according to the prediction result. The pet eating prediction model is trained based on pet eating statistical data of multiple second pets in the pet community.

2. The method according to claim 1, characterized in that The training process of the pet eating prediction model includes: Obtaining second pet attribute information, second pet behavior sequence information, and pet eating statistics of each of the second pets in the pet community, where the pet eating statistics and the second pet behavior sequence information are collected in the same time period; The second pet attribute information of the multiple second pets and the second pet behavior sequence information are used as sample inputs, and the pet eating statistics of the corresponding second pets are used as sample labels to adjust the parameters of the pre-trained neural network model to obtain a pet eating prediction model.

3. The method according to claim 2, characterized in that The method uses the second pet attribute information and the second pet behavior sequence information of the plurality of second pets as sample inputs, and uses the pet eating statistics of the corresponding second pets as sample labels to adjust the parameters of the pre-trained neural network model to obtain a pet eating prediction model, including: Based on the second pet attribute information of the plurality of second pets and the corresponding second pet behavior sequence information, a pre-trained neural network model is called to predict the pet eating habits, and a plurality of pet eating prediction data corresponding to the plurality of second pets is obtained; constructing a loss function based on the plurality of pet eating prediction data and the pet eating statistical data of the plurality of second pets; The model parameters of the neural network model are iteratively updated for multiple rounds according to the loss function, and a pet eating prediction model is determined based on the model parameters when a preset convergence condition is reached.

4. The method according to claim 1, wherein The training process of the pet eating prediction model includes: Obtaining second pet attribute information, second pet behavior sequence information, first weather information, and pet eating statistics of each of the second pets in the pet community, wherein the pet eating statistics, the first weather information, and the second pet behavior sequence information are collected in the same time period; Using the second pet attribute information of the plurality of second pets, the first weather information, and the second pet behavior sequence information as sample inputs, and using the pet eating statistics of the corresponding second pets as sample labels, the parameters of the pre-trained neural network model are adjusted to obtain a pet eating prediction model; The pet eating prediction model based on the first pet attribute information and the first pet behavior sequence information performs pet eating prediction, and generates a pet feeding plan for the first pet according to the prediction result, including: The second weather information is obtained, and the pet eating prediction model is called to perform pet eating prediction based on the second weather information, the first pet attribute information and the first pet behavior sequence information, and a pet feeding plan for the first pet is generated according to the prediction result.

5. The method according to claim 1, wherein The receiving of a pet feeding plan generating instruction for a first pet, and acquiring first pet attribute information and first pet behavior sequence information of the first pet according to the pet feeding plan generating instruction, includes: receiving a pet feeding plan generation instruction for a first pet, obtaining first time period information of a feeding plan application from the pet feeding plan generation instruction, and generating pet attribute information of the first pet based on the first time period information; Obtain first pet behavior sequence information of the first pet within a time range corresponding to a preset second time period information.

6. The method according to any one of claims 1 to 5, characterized in that The process of generating the pet feeding plan generation instruction for the first pet includes: Display the application interface of the pet feeding equipment management application; In response to a trigger operation received in the application interface to create a feeding plan for a first pet, displaying a pet feeding plan creation interface for the first pet, the pet feeding plan creation interface including an automatic creation control; In response to a triggering operation on the automatic creation control, a pet feeding plan generation instruction for the first pet is generated.

7. The method according to claim 6, characterized in that After calling the pet eating prediction model based on the first pet attribute information and the first pet behavior sequence information to predict the pet eating, and generating a pet feeding plan for the first pet according to the prediction result, the method further includes: Displaying a pet feeding plan generation prompt on the application interface; In response to a triggering operation for generating a prompt for the pet feeding plan, displaying a display page for the pet feeding plan, the display page including a feeding plan application control; In response to a triggering operation of the feeding plan application control, the pet feeding plan is applied to the feeding device.

8. A pet feeding plan processing device, characterized in that: The pet feeding plan processing device includes: a receiving unit, configured to receive a pet feeding plan generation instruction for a first pet, and acquire first pet attribute information and first pet behavior sequence information of the first pet according to the pet feeding plan generation instruction; a determining unit, configured to determine the pet community where the first pet is located according to the first pet attribute information; A generation unit is used to call a pet eating prediction model corresponding to the pet community to perform pet eating prediction based on the first pet attribute information and the first pet behavior sequence information, and generate a pet feeding plan for the first pet according to the prediction result, wherein the pet eating prediction model is trained based on pet eating statistical data of multiple second pets in the pet community.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the pet feeding plan processing method according to any one of claims 1 to 8 is implemented.

10. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the pet feeding plan processing method according to any one of claims 1 to 8 is implemented.

11. A computer program product comprising a computer program, characterized in that The computer program is read and executed by a processor of a computer device, so that the computer device executes the pet feeding plan processing method according to any one of claims 1 to 8.

Images