Poultry breeding index evaluation method, device and equipment and computer readable storage medium

Abstract

The invention provides a poultry breeding index evaluation method, apparatus and device, and a computer readable storage medium. The method comprises the steps of obtaining feature data of poultry; constructing a neural network model, and training the neural network model to obtain a breeding index evaluation model; and inputting the feature data into the breeding index evaluation model, and outputting a poultry breeding index evaluation result. In the scheme provided by the embodiment of the invention, the feature data of the poultry is predicted by using the neural network model, automatic evaluation of the poultry breeding index is realized, and a large amount of labor cost and time can be saved. For large-scale breeding scenes such as farms and the like, support is quickly and accurately provided for poultry breeding index evaluation decision, excellent poultry can be more quickly screened out, and the breeding success rate, efficiency and quality of the poultry are improved.

Description

Technical Field

[0001] The present application relates to the technical field of egg-laying poultry breeding, and in particular to a poultry breeding index evaluation method, device, equipment and computer-readable storage medium. Background Art

[0002] With the steady development of social economy, under the guidance of Internet of Things technology, in order to achieve fine breeding and management as well as traceability and management of breeding products, intelligent breeding has become a trend of sustainable development of the breeding industry.

[0003] Traditional breeding methods mainly rely on the experience and observation of breeders, which may lead to subjectivity and individual differences, and are not scientific and accurate enough. At the same time, they require a lot of time and manpower, which limits the efficiency and speed of breeding.

[0004] On the other hand, a lot of poultry data are not currently integrated together. The dispersion and lack of integration of data may lead to information islands, limiting the comprehensive analysis and optimization of the entire breeding process. Summary of the invention

[0005] Multiple aspects of the present application provide a poultry breeding index evaluation method, device, equipment and computer-readable storage medium, which realizes the automated evaluation of poultry breeding indicators by predicting the characteristic data of poultry using a neural network model, thereby solving the technical problems of low success rate and efficiency of poultry breeding in the prior art.

[0006] In order to achieve the above technical effects, one aspect of the present application provides a poultry breeding index evaluation method, comprising:

[0007] Obtain characteristic data of birds;

[0008] Constructing a neural network model, and training the neural network model to obtain a breeding index evaluation model;

[0009] The characteristic data is input into the breeding index evaluation model, and the poultry breeding index evaluation result is output.

[0010] According to a preferred embodiment of the present invention, the step of obtaining characteristic data of poultry further comprises:

[0011] Obtaining data on the physiological characteristics, egg-laying data, and hatching data of poultry;

[0012] The eggs and hatching birds are marked separately and the data are linked to the laying birds;

[0013] Clean the acquired data and remove unqualified data;

[0014] Convert qualified data into feature data.

[0015] According to a preferred embodiment of the present invention, the step of obtaining the physiological characteristic data, egg laying data and hatching data of poultry further comprises:

[0016] The physiological characteristic data of the poultry include: weight, age, body shape and genetic data of the poultry;

[0017] The egg-laying data include: egg-laying quantity and egg size, shape, and color data;

[0018] The hatching data include: hatched poultry weight, feather color, toe number and beak shape data.

[0019] According to a preferred embodiment of the present invention, the step of constructing a neural network model and training the neural network model to obtain a breeding index evaluation model further comprises:

[0020] Determine the number of layers of the neural network, the number of neurons in each layer, and the activation function;

[0021] Compile the model, specify the loss function, optimizer, and evaluation metric;

[0022] The historical data are used as samples to train the model, adjust the model parameters, and finally obtain the breeding index evaluation model.

[0023] According to a preferred embodiment of the present invention, the method further comprises: evaluating the breeding index evaluation model, further comprising:

[0024] Setting a test set and a validation set, and inputting the test set into the breeding index evaluation model to obtain the prediction result of the model;

[0025] Set and calculate evaluation indicators based on the prediction results;

[0026] The obtained prediction results and evaluation indicators are compared with the validation set to evaluate the performance of the breeding indicator evaluation model.

[0027] According to a preferred embodiment of the present invention, the step of inputting the characteristic data into the breeding index evaluation model and outputting the poultry breeding index evaluation result further comprises:

[0028] Inputting the characteristic data of the poultry to be tested into the breeding index evaluation model to obtain the probability that the poultry breeding index is qualified;

[0029] Set a probability threshold, when the probability is greater than the threshold, it is qualified, and when the probability is less than the threshold, it is unqualified;

[0030] Unqualified poultry are screened out based on the threshold and treated accordingly.

[0031] According to a preferred embodiment of the present invention, the method further comprises:

[0032] The training samples are updated regularly, and the breeding index evaluation model parameters are updated according to the training samples.

[0033] Another aspect of the present application provides a poultry breeding index evaluation device, comprising:

[0034] A data acquisition module, used for acquiring characteristic data of poultry;

[0035] A model training module is used to construct a neural network model and train the neural network model to obtain a breeding index evaluation model;

[0036] The breeding index evaluation module is used to input the characteristic data into the breeding index evaluation model and output the poultry breeding index evaluation results.

[0037] Another aspect of the present application provides an electronic device for evaluating poultry breeding indicators, the device comprising:

[0038] at least one processor; and

[0039] a memory communicatively connected to the at least one processor; wherein,

[0040] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described above.

[0041] In another aspect of the present application, a computer-readable storage medium is provided, on which computer program instructions are stored. The computer program instructions can be executed by a processor to implement the above method.

[0042] In the solution provided in the embodiment of the present application, a neural network model is used to predict the characteristic data of poultry and realize the automated evaluation of poultry breeding indicators, which can save a lot of manpower costs and time. For large-scale breeding scenarios such as farms, it can quickly and accurately provide support for poultry breeding indicator evaluation decisions, and can more quickly screen out excellent poultry, thereby improving the breeding success rate, efficiency and quality of poultry. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0044] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0045] Figure 1 A flowchart of a poultry breeding index evaluation method provided in one embodiment of the present application;

[0046] Figure 2 A schematic diagram of a poultry breeding index evaluation device provided in one embodiment of the present application;

[0047] Figure 3 The present invention is a schematic diagram of the structure of a device suitable for implementing the solution in the embodiment of the present application.

[0048] The same or similar reference numerals in the drawings represent the same or similar components. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0050] In a typical configuration of the present application, the terminal and the equipment of the service network each include one or more processors (CPU), input / output interface, network interface and memory.

[0051] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0052] Computer readable media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer program instructions, data structures, modules of programs or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0053] In actual scenarios, the execution subject of the method can be a user device, or a device formed by integrating a user device and a network device through a network, or an application running on the above device. The user device includes but is not limited to various terminal devices such as computers, mobile phones, tablet computers, smart watches, and bracelets. The network device includes but is not limited to network hosts, single network servers, multiple network server sets, or cloud computing-based computer sets, which can be used to implement some processing functions when setting an alarm. Here, the cloud is composed of a large number of hosts or network servers based on cloud computing, where cloud computing is a type of distributed computing, a virtual computer composed of a group of loosely coupled computer sets.

[0054] In order to realize the scheme of the present invention, the present invention provides a poultry breeding index evaluation method, device, equipment and computer-readable storage medium. The neural network model is used to predict the characteristic data of poultry, screen out the information directly related to the breeding index, realize the automatic evaluation of poultry breeding index, improve the breeding efficiency and success rate, and help promote the digital and intelligent development of the agricultural industry.

[0055] The poultry described in the present invention include but are not limited to common poultry such as chickens, ducks, and geese, among which the female poultry focuses on its reproductive indicators, which include but are not limited to physiological characteristics, egg-laying data, and hatching data. The embodiment of the present invention is described using hens among female poultry as an example, and other poultry may also be applicable to the method of the present invention.

[0056] Figure 1 A flowchart of a poultry breeding index evaluation method provided in one embodiment of the present application is shown in FIG. Figure 1 As shown, the poultry breeding index evaluation method comprises at least the following steps:

[0057] S101. Obtain characteristic data of poultry.

[0058] Specifically, S1011, obtaining physiological characteristic data, age data, egg data and hatching data of the hen, wherein the physiological data includes:

[0059] (1) Weight: The weight of hens is an important physiological characteristic, positively correlated with body shape. Under normal circumstances, it is expressed as an S-shaped curve growth of weight, usually in kilograms (kg), for example: 2.5kg. In addition, weight is also related to bones, feed consumption rate, reproductive performance, and abdominal fat.

[0060] (2) Body Condition Score: used to assess whether the hen’s body condition is healthy, for example: 3.5 (usually a score from 1 to 5, with 5 indicating the best body condition).

[0061] (3) Gene expression: The genetic data of hens may include the expression of some important genes, such as the following genes and their expression:

[0062] IGF-1 (insulin-like growth factor-1): 0.8 (high level expression), usually associated with growth and muscle growth.

[0063] TLR (Toll-like receptor): 0.6 (moderate level of expression), involved in the regulation of the immune system and the process of resisting diseases.

[0064] FSH (Follicle Stimulating Hormone): 0.8 (high level of expression), a hormone that regulates the reproductive system. A high level of expression means higher reproductive capacity.

[0065] LH (luteinizing hormone): 0.2 (moderate level of expression). It is also a hormone that regulates the reproductive system. Moderate levels of expression may indicate that reproductive function is within a moderate range.

[0066] Age data: The age of the hen is also a key characteristic, usually in days, for example: 30 days.

[0067] Egg data include:

[0068] (1) Egg Production: The number of eggs laid by a hen per day or per year, for example: 200 eggs per year;

[0069] (2) Egg weight and size are usually measured in grams (g) and millimeters (mm), for example, 60 grams per egg. They are mainly affected by the age of the laying hen, but are also related to body weight, age at first laying, location of consecutive laying, nutritional level, air temperature, light exposure time, temperature, disease, etc.

[0070] (3) Egg Shape: The external shape of the egg, which may be long, round, etc.

[0071] (4) Eggshell Color: The color of the eggshell may be white, brown, or brown, for example, brown. Brown-shelled eggs have a higher heritability.

[0072] (5) Eggshell quality: Eggshell strength and whether there are deformities, cracks, brown spots on the eggshell surface are also important indicators of poultry eggs.

[0073] (6) Protein ratio: The height of concentrated protein (Haugh unit) is the main indicator for determining protein quality.

[0074] Incubation data includes:

[0075] (1) Chick Weight: The weight of a newly hatched chick, usually in grams (g). For example: 40g;

[0076] (2) Chick Feather Color: The color of the feathers of newly hatched chicks, which may be yellow, black, etc. For example: White feathers: dominant white feathers IICC, recessive white feathers iicc; Black feathers: horizontal stripes (B / b); light flowers; Yellow feathers; Gold and silver feathers (S / s) The feather color is determined by the sex;

[0077] (3) Chick Toe Count: The number of toes on a chick, usually two toes, three toes, etc. For example: three toes;

[0078] (4) Chick Beak Shape: The shape of the chick's beak, which may be sharp, flat, etc. For example: sharp.

[0079] S1012. Mark the eggs and hatched birds separately, and associate the data with the egg-laying birds.

[0080] After obtaining the physiological characteristics data, egg-laying data, and hatching data of the hen, the data is associated. First, the hatched eggs are marked with invisible ink. When the hen lays eggs, the eggs will be sprayed with invisible ink to form a unique identification ID. This ID can be a combination of numbers, letters, or symbols to ensure that each egg is unique. When the ID of the egg needs to be read, ultraviolet light can be used to illuminate the invisible ink on the eggshell. Through the appropriate ultraviolet irradiation equipment, the invisible ink can be made visible under ultraviolet light, so that the ID of the egg can be read.

[0081] After the egg is sprayed with invisible ink, this ID is linked to the hen's information, such as the hen's ID (which can be obtained through a wearable device), weight, genetic data, etc.

[0082] This association can be recorded and stored through a relational database (such as MySQL, PostgreSQL, SQL Server, etc.) to ensure that the ID of each egg and the information of its corresponding hen are correctly associated.

[0083] When the chicks are hatched from the eggs, the egg ID can also be used to associate with the chick's information (such as the chick's ID (obtained through wearable devices), phenotypic data, etc.), and then with the hen's information to ensure the correlation between the eggs, hens and chicks.

[0084] In this way, by reading the ID of the egg and querying the associated database, the information of the hatched chicks and the hen can be obtained, and the data of the eggs, chicks and hens can be associated.

[0085] S1013. Clean the acquired data and remove unqualified data.

[0086] Data cleaning, processing of missing values ​​and outliers, deduplication, etc.

[0087] Specifically, (i) Processing missing values: Detect and process missing values ​​in the data. You can delete rows containing missing values, or fill in missing values ​​(mean, median, mode, etc.) based on the distribution of the data.

[0088] (1) Mean: When the data presents an approximate normal distribution (or symmetrical distribution), it is appropriate to use mean filling. Among them, the physiological characteristics data of hens, egg production data, and phenotypic data of chicks hatched from eggs can be processed with the mean. The formula is:

[0089]

[0090] Among them, X i : The specific observation value of a hen or a batch of eggs (the value that needs to be filled);

[0091] The mean of the corresponding feature in the entire dataset;

[0092] Take the weight characteristic of hens as an example. If there are missing values ​​in the weight data;

[0093] X i : The observed weight of a hen;

[0094] The mean weight of the entire data set;

[0095] Mean filling formula: Mean filling formula:

[0096] (2) Median: When there are many outliers in the data, using median filling is more robust because the median is not affected by outliers.

[0097] Median filling is usually used to deal with situations where there are many outliers in the data set. When processing physiological characteristic data of hens, egg laying data, and phenotypic data of chicks hatched from eggs, if there are many outliers in some characteristic data, you can consider using median filling.

[0098] The specific processing steps are as follows:

[0099] Detect outliers: First, you need to perform outlier detection on each feature. You can use statistical methods such as Z-score to identify outliers (described in detail in the next section).

[0100] Calculate the median: For each feature, calculate its median.

[0101] Fill missing values: fill missing values ​​in features with the median of the corresponding features.

[0102] For example, if there is a column of hen weight data, and there are many outliers in this data, you can handle it like this:

[0103] Detect outliers: Calculate the Z-score of weight and mark values ​​greater than a threshold as outliers.

[0104] Calculate Median: Calculate the median of the weight data.

[0105] Fill missing values: Fill missing values ​​in weight data with the calculated median.

[0106] (3) Mode: Mode filling is usually used to process discrete numerical features or categorical variables. If some feature data is discrete or has categorical properties, you can consider using mode filling.

[0107] The specific processing steps are as follows:

[0108] Detecting discrete or categorical variables: First, we need to determine which features are discrete or categorical in nature.

[0109] Calculate the mode: For each discrete feature, calculate its mode, which is the value with the highest frequency in the data set.

[0110] Fill missing values: fill missing values ​​in a feature with the mode of the corresponding feature.

[0111] Suppose there is a column of genetic data:

[0112] 1.AA

[0113] 2.AG

[0114] 3.GG

[0115] 4.AG

[0116] 5.NaN

[0117] 6.AA

[0118] 7.GG

[0119] 8.NaN

[0120] 9.AG

[0121] 10.GG

[0122] Among them, the genetic data here includes three genotypes: AA: represents that the two alleles of the individual at this gene site are both type A; AG: represents that the two alleles of the individual at this gene site are one type A and one type G; GG: represents that the two alleles of the individual at this gene site are both type G; missing values ​​(NaN) indicate that the genotype data of these samples at this gene site are missing and need to be processed according to the method of filling missing values.

[0123] First, determine that the genetic data is discrete and has categorical properties. Then, calculate the mode of the genetic data. In this example, the mode is AG because it appears most frequently in the data set. Finally, fill the missing values ​​(NaN) with the mode AG, and get the following results:

[0124] 1.AA

[0125] 2.AG

[0126] 3.GG

[0127] 4.AG

[0128] 5.AG

[0129] 6.AA

[0130] 7.GG

[0131] 8.AG

[0132] 9.AG

[0133] 10.GG

[0134] (ii) Handling outliers: Detect and handle outliers. Statistical methods (such as Z-score) can be used to determine whether it is an outlier and correct or delete it.

[0135] The following will illustrate this process with the existing data:

[0136] (1) Assume that there is weight data of hens:

[0137] 1.2.5kg

[0138] 2.2.7kg

[0139] 3.2.1kg

[0140] 4.2.9kg

[0141] 5.2.2kg

[0142] 6.2.5kg

[0143] 7.2.2kg

[0144] 8.2.8kg

[0145] 9.2.0kg

[0146] 10.2.1kg

[0147] (2) Calculate Z-score:

[0148] First, calculate the Z-score of the hen's weight. Assume that the mean (μ) and standard deviation (σ) of the weight have been calculated. Now calculate the Z-score of each sample using the formula:

[0149] Where X is the weight of each sample, μ is the mean weight, and μ is the standard deviation of weight.

[0150] The Z-score of each sample is calculated as follows:

[0151] For a 2.5kg hen, the Z-score is

[0152] For a 2.7kg hen, the Z-score is

[0153] And so on, calculate the Z-score of all samples.

[0154] (3) Define the threshold: Based on business requirements and statistical principles, select a threshold to determine what is an outlier. Here, a Z-score greater than 3 or less than -3 is selected as the outlier threshold.

[0155] (4) Mark outliers: For each sample, compare whether its Z-score is greater than 3 or less than -3. If this condition is met, mark it as an outlier.

[0156] (5) Handling outliers: Depending on the specific situation, the following strategies can be adopted to handle outliers:

[0157] Remove outliers: directly remove samples marked as outliers.

[0158] Replace outliers: Replace outliers with the mean, median, or other suitable value.

[0159] (iii) Remove duplicate data: Remove duplicate data rows to avoid duplicate samples causing bias in model training.

[0160] (1) Detecting duplicate data rows: First, check the data set to find out whether there are exactly the same samples. These samples may appear repeatedly in the data set due to data collection or other reasons.

[0161] (2) Marking duplicate data rows: For identical samples found, one or more of them can be marked as duplicate rows. This can be achieved by comparing the rows and marking the duplicates.

[0162] (3) Remove duplicate data rows: Once the duplicate samples are marked, the following actions can be taken depending on the specific situation:

[0163] Delete duplicate rows: directly delete the rows marked as duplicates and keep one of the samples.

[0164] Keep one copy: You can also choose to keep one copy of all duplicate rows and delete the others.

[0165] For example, there are the following physiological characteristics data of hens:

[0166] 1. Weight: 2.5kg, Age: 548 days, Number of eggs laid: 200 eggs / year, Genetic data: [0.8, 0.6, -0.3, 0.2]

[0167] 2. Weight: 3.0kg, Age: 730 days, Number of eggs laid: 180 eggs / year, Genetic data: [0.7, 0.5, -0.2, 0.1]

[0168] 3. Weight: 2.8kg, Age: 657 days, Number of eggs laid: 190 eggs / year, Genetic data: [0.9, 0.7, -0.4, 0.3]

[0169] 4. Weight: 2.5kg, Age: 548 days, Number of eggs laid: 200 eggs / year, Genetic data: [0.8, 0.6, -0.3, 0.2] (repeated rows)

[0170] Items 1 and 4 are duplicates. Therefore, you can mark these duplicates and delete them or keep one of the copies as needed.

[0171] S1014. Convert qualified data into feature data.

[0172] First, feature selection is performed to select features that are meaningful to the model and exclude features that are not helpful to the model.

[0173] Tree-based methods (such as random forests): Tree models can calculate the importance of features during training and select features based on their importance.

[0174] Specifically: Prepare the physiological characteristics data of the hen, the egg-laying data, and the phenotypic data of the chicks hatched from the eggs as inputs to the model.

[0175] Choose appropriate parameters, such as the number of decision trees (n_estimators) and the maximum depth of each tree (max_depth).

[0176] For the number of decision trees (n_estimators), you can start with a smaller value, such as 10, and then gradually increase the number to observe the model performance.

[0177] For the depth of the tree (max_depth), you can also start with a smaller depth, such as 3 or 5, and then gradually increase the depth based on the performance of the validation set.

[0178] Then the feature importance is calculated. Each decision tree in the random forest can calculate the importance of the feature.

[0179] We choose to use Mean Decrease Accuracy as the metric, which is to calculate the accuracy reduction of each feature on each decision tree in the random forest and then take the average. After calculating the importance score of each feature, we can normalize it for comparison and sorting.

[0180] Normalization can make the sum of the importance scores of all features equal to 1, or scale them proportionally.

[0181] Finally, the features are sorted according to their importance scores, with features with higher importance at the front and features with lower importance at the back.

[0182] After selecting appropriate features, the physiological characteristic data of the hen, the egg-laying data, and the phenotypic data of the chicks hatched from the eggs are converted into numerical data that can be processed by the model, and the numerical features are normalized to ensure the stability and effectiveness of the model training.

[0183] Specifically, for non-numeric data, such as categorical variables, One-Hot Encoding can be used to convert them into numeric data.

[0184] For example, if there is a hen breed feature that may have multiple categories, such as "A", "B", and "C", this feature can be converted into three new binary features, representing the presence or absence of each category. For example, "A" may be encoded as 1,0,01,0,0, "B" as 0,1,00,1,0, and "C" as 0,0,10,0,1.

[0185] For text data, text processing methods (such as bag-of-words models or word embeddings) can be used to convert it into numerical features that the model can process.

[0186] Data normalization is to scale numerical features to a standard range to eliminate the dimensional effects between different features, making the model more stable and converging faster.

[0187] Normalization is to scale the numerical features to the range of 0, 10, 1. The formula is as follows:

[0188]

[0189] This ensures that the value range of all features is between 0, 10, and 1, preventing some features from having too much impact on model training.

[0190] Standardization is to convert numerical features into a standard normal distribution with a mean of 0 and a standard deviation of 1. The formula is as follows:

[0191]

[0192] This method is suitable for the case where the distribution of features is approximately normal, and can make the value ranges of different features less different.

[0193] Combining the above steps and sample data, the following processing can be performed:

[0194] (1) For the physiological characteristics data (such as body weight) and age characteristics data of hens:

[0195] If there are non-numeric features, such as species, they can be one-hot encoded to convert them into binary form.

[0196] For numerical features, normalization or standardization can be performed so that the value range of all features is 0, 10, 1 or conforms to the standard normal distribution.

[0197] (2) For poultry egg data, such as egg production, egg size, shape, eggshell color, eggshell quality, egg white ratio, etc.:

[0198] If these data are numerical, they can also be normalized or standardized.

[0199] (3) Phenotypic data of chicks hatched from eggs, such as growth rate of chicks, hatching rate of eggs, etc.:

[0200] Normalization or standardization can also be performed so that all features are within a unified value range.

[0201] For example, for the gene data 0.8, 0.6, -0.3, 0.20.8, 0.6, -0.3, 0.2, you can:

[0202] If these gene data represent the expression level of specific genes, they can be regarded as numerical data and can be normalized or standardized.

[0203] If these genetic data represent genotypes (for example, 0 represents one gene and 1 represents another gene), you can consider using one-hot encoding to convert them into binary form.

[0204] S102, constructing a neural network model, and training the neural network model to obtain a breeding index evaluation model.

[0205] First, define the model structure. Specifically,

[0206] Determine the number of layers of the neural network and the number of neurons and activation functions in each layer.

[0207] (1) Input layer: The feature dimension is the dimension of the physiological characteristic data. For example, the physiological characteristic data of a hen is used as input, and the feature dimension is 10.

[0208] (2) Hidden layers: The first and second hidden layers each contain 64 neurons. This number is a common choice and can be adjusted based on specific problems and experimental results.

[0209] Use ReLU activation function: ReLU is a commonly used activation function in deep learning that can help the network learn nonlinear relationships.

[0210] (3) Output layer: The output layer contains 1 neuron, corresponding to the output of the binary classification problem (0 or 1).

[0211] Use Sigmoid activation function: The Sigmoid function can compress the output value to between 0 and 1, which can represent the probability that the sample belongs to a certain category.

[0212] After defining the model structure, you need to compile the model and specify the loss function, optimizer, and evaluation metrics. Specifically,

[0213] Select loss function: Select an appropriate loss function according to the task type, such as the cross-entropy loss function.

[0214] The formula is:

[0215]

[0216] in:

[0217] N is the sample size;

[0218] y i is the true label of the i-th sample, which takes a value of 0 or 1;

[0219] p i is the model's predicted probability that the i-th sample belongs to category 1.

[0220] For each sample, the first term y i log(p i ) is the loss when the true label is 1, which means the loss when the model predicts that it is class 1 and the actual class is class 1. If y i is 1, then the value of this term is log(p i ); if y i If it is 0, the value of this item is 0.

[0221] The second term (1-y i )log(1-p i ) is the loss when the true label is 0, which means the loss when the model predicts that the class is 0 and the actual class is 0. If y i is 0, then the value of this term is log(1-p i ); if y i is 1, the value of this item is 0.

[0222] Therefore, when the true label is 1, we want the model to predict a probability p i The closer it is to 1, the smaller the loss;

[0223] When the true label is 0, we want the model to predict the probability p i The closer it is to 0, the smaller the loss.

[0224] Select an optimizer to minimize the loss function. Select an optimizer such as Adam according to your needs.

[0225] (1) Calculate the gradient of the loss function with respect to the parameters: Derivative the loss function and calculate the gradient of each parameter. Taking weight W as an example, the gradient calculation formula is as follows:

[0226]

[0227] For each parameter in the neural network, its gradient with respect to the loss function is calculated.

[0228] (2) Update model parameters: Use the update rule of the Adam optimizer to update the model parameters. The Adam optimizer uses the first-order moment estimate of the gradient (i.e., the mean value of the gradient) and the second-order moment estimate (i.e., the mean square value of the gradient) to adjust the learning rate.

[0229] Adam's parameter update formula is as follows:

[0230]

[0231] Among them: θ is the parameter to be updated;

[0232] is the gradient of the loss function with respect to the parameter θ;

[0233] α is the learning rate;

[0234] m is the first-order moment estimate (mean) of the gradient, initialized to 0;

[0235] ν is the second-order moment estimate of the gradient (mean square value), initialized to 0;

[0236] β1 and β2 are the decay rates, usually taking the default values ​​of β1 = 0.9 and β2 = 0.999;

[0237] ∈ is a small constant added for numerical stability, usually 10 -8 .

[0238] At each training step, the value of each parameter θ is updated using the above formula.

[0239] Repeatedly select the loss function and minimize the loss function, and continue to iterate and update the parameters until the specified number of training rounds is reached or the loss function converges.

[0240] Finally, add evaluation indicators: If you need to monitor the performance indicators of the model during training, you can add some evaluation indicators, such as accuracy.

[0241] When training a model, you can define a function to calculate the accuracy. The accuracy is calculated as:

[0242]

[0243] The accuracy can be calculated once after each epoch to understand the performance of the model on the training set.

[0244] After adding indicators to the model, the model is trained. Before training, the training data needs to be prepared. This includes input features and corresponding labels. Historical data can be used as samples. During the training process, some training parameters need to be set, such as the number of training rounds (epochs) and the number of samples in each batch (batch_size). When the training is completed, the trained model can be saved as a file for subsequent prediction and deployment.

[0245] Preferably, after the training is completed to obtain the breeding index evaluation model, the breeding index evaluation model is tuned and evaluated, and a test set and a validation set are set.

[0246] First, tune the model, use the validation set to verify the model, and adjust the hyperparameters, such as the number of hidden layers, the number of neurons in each layer, and the learning rate, to optimize the model performance.

[0247] Specifically, during the training process, a portion of the data is usually used as a validation set to evaluate the performance of the model on unseen data. This validation set should be independent of the training set and cannot be used for training.

[0248] Select the hyperparameters that need to be adjusted based on the characteristics of the existing data and model. For example, in the physiological characteristics of hens, egg-laying data, and phenotypic data of chicks hatched from eggs, you can consider adjusting the number of hidden layers and the number of neurons in each layer of the neural network model.

[0249] Then establish the parameter search space: for each hyperparameter to be adjusted, a search range or candidate set needs to be determined. For example, the number of hidden layers can be selected between 1 and 3, the number of neurons in each layer can be between 32 and 128, the learning rate can be selected between 0.001 and 0.01, and so on.

[0250] Tuning process: For each combination of hyperparameters, perform the following steps:

[0251] Train model: Use the training set to train the model according to the current hyperparameter settings.

[0252] Validation model: Use the validation set to evaluate the performance of the model, such as calculating the loss value and accuracy on the validation set.

[0253] Record results: Record the validation results of each set of hyperparameters. You can choose to save the best model and the corresponding hyperparameters.

[0254] Adjust hyperparameters: According to the verification results, adjust the value range of hyperparameters and reset the parameter combination for a new round.

[0255] Loop iteration: Repeat the above steps until the preset number of iterations is reached or the optimal hyperparameter combination is found.

[0256] Finally, select the best model: After the tuning is completed, select the best performing model and corresponding hyperparameter combination based on the performance of the validation set.

[0257] After the tuning is completed, the model is evaluated. Specifically, a test set is prepared first. This test set is data that has never been used in the training and validation process. This ensures that the evaluation result is a true reflection of the model's generalization ability.

[0258] These data should include:

[0259] Physiological characteristics of hens: such as weight, body length, age, etc.

[0260] Egg production data: including egg production, eggshell color, etc.

[0261] Phenotypic data of chicks hatched from eggs: such as growth rate, feather color, etc.

[0262] The data in the test set should not have appeared in the previous model building, training and validation process to ensure that the evaluation results are reliable for the generalization ability of the model.

[0263] During the tuning process, we saved the models with different hyperparameters. Now we need to load the model that performs best on the validation set. Use the loaded best model to predict the test set and evaluate the model performance. Input the features of the test set into the model to get the model's prediction results.

[0264] Then the breeding indicators of hens are classified and evaluated, so the evaluation indicators of classification problems can be used to evaluate the performance of the model on the test set. The precision, recall and F1-score are calculated based on the confusion matrix.

[0265] Suppose you want to predict whether a hen will lay eggs (a binary classification problem), you can define the following indicators:

[0266] Precision: The ratio of samples predicted to be egg-laying and actually laying eggs to all samples predicted to be egg-laying.

[0267] Recall: The ratio of samples that are actually laid and predicted to be laid to all samples that are actually laid.

[0268] F1-score: The harmonic mean of precision and recall, which comprehensively considers the accuracy and recall of the model.

[0269] Combine data to calculate indicators. For example, there is a set of data:

[0270]

[0271] Among them, positive example:

[0272] Hens with high egg production (Egg Production>200 eggs / year);

[0273] Hens with larger body weight (Weight>3kg);

[0274] Hens with good body condition score (Body Condition Score>3.5).

[0275] Highly expressed genes:

[0276] IGF-1 (insulin-like growth factor-1): 0.8 or higher;

[0277] FSH (follicle stimulating hormone): 0.8 or higher;

[0278] The chicks are heavier (Chick Weight>50g);

[0279] The eggs laid by hens are brown (Egg Color is brown);

[0280] The chick feather color is yellow (Chick Feather Color is yellow);

[0281] Chick Toe Count is three toes.

[0282] Negative example:

[0283] Hens with low egg production (Egg Production <= 200 eggs / year);

[0284] Hens with smaller body weight (Weight <= 3kg);

[0285] Hens with lower body condition scores (Body Condition Score <= 3.5).

[0286] Genes expressed at moderate or low levels:

[0287] IGF-1 (insulin-like growth factor-1): less than 0.8;

[0288] FSH (follicle stimulating hormone): less than 0.8;

[0289] The chicken weight is small (Chick Weight <= 50g);

[0290] The eggs laid by hens are white (Egg Color is white);

[0291] The chick feather color is black (Chick Feather Color is black);

[0292] The number of toes of a chick is two.

[0293] First, calculate the confusion matrix:

[0294] True Positive (TP) = 4 (samples 1, 3, 6, and 7 are positive examples, and the model predicts correctly): the number of samples that are predicted to be positive examples and are actually positive examples.

[0295] False Positive (FP) = 1 (sample 2 is a negative example, but is mistakenly predicted as a positive example): the number of samples predicted as positive examples but actually negative examples.

[0296] False Negative (FN) = 2 (samples 5 and 8 are negative examples, but are mistakenly predicted as positive examples): the number of samples predicted as negative examples but actually positive examples.

[0297] True Negative (TN) = 1 (sample 4 is a negative example, and the model predicts correctly): the number of samples that are predicted to be negative examples and are actually negative examples.

[0298] The calculation formulas for precision, recall and F1-score are:

[0299] Precision

[0300] Precision refers to the proportion of positive examples predicted by the model to those that are actually positive examples. The calculation formula is:

[0301]

[0302] Recall

[0303] Recall refers to the proportion of samples correctly predicted by the model as positive examples to all true positive examples. The calculation formula is:

[0304]

[0305] F1-score

[0306] F1-score is the harmonic mean of precision and recall, which combines the accuracy and recall of the model. The calculation formula is:

[0307]

[0308] According to the above formula, we can get:

[0309] Precision = TP / (TP+FP) = 4 / (4+1) = 0.8

[0310] Recall = TP / (TP+FN) = 4 / (4+2) = 0.67

[0311] F1-score=2*(Precision*Recall) / (Precision+Recall)=2*(0.8*0.67) / (0.8+0.67)=0.73

[0312] Therefore, the F1-score calculated based on these data is 0.73.

[0313] Finally, the evaluation indicators and results are output to analyze the performance of the model. These results can be compared with the results on the validation set to confirm the generalization ability of the model on unknown data.

[0314] S103, inputting the characteristic data into the breeding index evaluation model, and outputting the poultry breeding index evaluation result.

[0315] By using the trained neural network model and inputting the characteristic data of new hens, the predicted breeding indicators can be obtained.

[0316] Specifically, first, load the previously trained neural network model to ensure that the model parameters and structure are correct. Then prepare the new hen's feature data: ensure that the input feature data format is consistent with the model's input requirements.

[0317] Preprocess the characteristic data of the new hens, such as normalization and standardization, to ensure that the data processing method is consistent with that during training.

[0318] For example, there is a new hen data, and we hope to predict its breeding index through the characteristic data of this hen. The characteristic data of this hen are as follows:

[0319] Feature 1: Weight: 2.5kg

[0320] Feature 2: Age: 550 days

[0321] Feature 3: Body Condition Score: 3.5

[0322] Feature 4: Genetic data:

[0323] IGF-1 (Insulin-like Growth Factor-1): 0.8

[0324] TLR (Toll-like receptor): 0.6

[0325] FSH (Follicle Stimulating Hormone): 0.8

[0326] LH (luteinizing hormone): 0.2

[0327] Feature 5: Egg laying data:

[0328] Egg Production: 200 eggs / year

[0329] Egg Size: 60g / piece

[0330] Egg Shape: Long

[0331] Egg Color: Brown

[0332] Feature 6: Phenotypic data of chicks:

[0333] Chicken Weight: 40g

[0334] Chick Feather Color: Yellow

[0335] Chick Toe Count: Three

[0336] Chick Beak Shape: Pointed.

[0337] Then substitute this data into the model for prediction. Assuming that the model has been trained and loaded into memory, input new data into the model, and the model will give a probability value, indicating the probability that the hen belongs to the positive class (high breeding index).

[0338] Set threshold: Set a threshold to judge the prediction results of the model. When the probability is greater than the threshold, it is qualified, and when the probability is less than the threshold, it is unqualified. For example, if the threshold is set to 0.5, the prediction results with a prediction value greater than 0.5 are considered positive, and the prediction value less than or equal to 0.5 is considered negative.

[0339] Model prediction: The physiological characteristics data, age data, egg data and hatching data of the hen are input into the model, and processed through the input layer, hidden layer and output layer in turn to obtain the qualified probability value.

[0340] Finally, the final prediction result is obtained: the output of the model is converted into the final classification result according to the threshold. If the probability of the model output is greater than the threshold, the prediction is that the breeding index is high; otherwise, the prediction is that the breeding index is low.

[0341] For example, the probability value output by the model is 0.65, which means that the model believes that the probability of the hen belonging to the positive class (high breeding index) is 65%. Then, based on the probability value output by the model and the set threshold, the final prediction result can be obtained: the model output probability is 0.65, which is greater than the threshold of 0.5, so the final prediction is that the breeding index is high.

[0342] For hens with a probability value lower than 0.5, it indicates that there is a problem with their health and further examination and timely treatment are required. If they are sick, they can be treated in time.

[0343] Preferably, in order to ensure the accuracy of the model, the training samples are updated regularly, the old training samples are replaced with recent ones, and the breeding index evaluation model parameters are updated according to the training samples.

[0344] Figure 2 A schematic diagram of a poultry breeding index evaluation device provided in one embodiment of the present application is shown in FIG. Figure 2 As shown, the device comprises:

[0345] A data acquisition module 11 is used to acquire characteristic data of poultry;

[0346] A model training module 22 is used to construct a neural network model and train the neural network model to obtain a breeding index evaluation model;

[0347] The breeding index evaluation module 33 is used to input the characteristic data into the breeding index evaluation model and output the poultry breeding index evaluation result.

[0348] The above-mentioned device can execute the poultry breeding index evaluation method in the above-mentioned embodiment, wherein the data acquisition module 11 executes step S101; the model training module 22 executes step S102; and the breeding index evaluation module 33 executes step S103.

[0349] Based on the same inventive concept, an electronic device is also provided in an embodiment of the present application, and the method corresponding to the electronic device may be the method for evaluating poultry breeding indicators in the aforementioned embodiment, and its principle of solving the problem is similar to that of the method. The electronic device provided in an embodiment of the present application includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the methods and / or technical solutions of the aforementioned multiple embodiments of the present application.

[0350] The electronic device may be a user device, or a device formed by integrating a user device and a network device through a network, or may be an application running on the above device. The user device includes but is not limited to various terminal devices such as computers, mobile phones, tablet computers, smart watches, and bracelets. The network device includes but is not limited to network hosts, single network servers, multiple network server sets, or cloud computing-based computer sets, which can be used to implement some processing functions when setting an alarm. Here, the cloud is composed of a large number of hosts or network servers based on cloud computing, where cloud computing is a type of distributed computing, a virtual computer composed of a group of loosely coupled computer sets.

[0351] Figure 3 The structure of a device suitable for implementing the method and / or technical solution in the embodiment of the present application is shown, and the device 1200 includes a central processing unit (CPU, Central Processing Unit) 1201, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM, Read Only Memory) 1202 or the program loaded from the storage part 1208 to the random access memory (RAM, Random Access Memory) 1203. In RAM1203, various programs and data required for system operation are also stored. CPU 1201, ROM 1202 and RAM 1203 are connected to each other through bus 1204. Input / output (I / O, Input / Output) interface 1205 is also connected to bus 1204.

[0352] The following components are connected to the I / O interface 1205: an input section 1206 including a keyboard, a mouse, a touch screen, a microphone, an infrared sensor, etc.; an output section 1207 including a cathode ray tube (CRT), a liquid crystal display (LCD), an LED display, an OLED display, etc., and a speaker, etc.; a storage section 1208 including one or more computer-readable media such as a hard disk, an optical disk, a magnetic disk, a semiconductor memory, etc.; and a communication section 1209 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 1209 performs communication processing via a network such as the Internet.

[0353] In particular, the methods and / or embodiments in the embodiments of the present application may be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes program code for executing the method shown in the flowchart. When the computer program is executed by the central processing unit (CPU) 1201, the above functions defined in the method of the present application are executed.

[0354] Another embodiment of the present application further provides a computer-readable storage medium having computer program instructions stored thereon, wherein the computer program instructions can be executed by a processor to implement the methods and / or technical solutions of any one or more embodiments of the present application described above.

[0355] Specifically, the present embodiment may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device, or device.

[0356] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than a computer readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0357] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0358] Computer program code for performing the operation of the present application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., using an Internet service provider to connect through the Internet).

[0359] The flow chart or block diagram in the accompanying drawings shows the possible architecture, function and operation of the equipment, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated system for hardware that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0360] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0361] In the several embodiments provided in the present application, 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 only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or page 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.

[0362] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0363] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0364] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform some steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.

[0365] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

[0366] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in a device claim can also be implemented by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

Claims

1. A poultry breeding index evaluation method, characterized in that: include: Obtain characteristic data of birds; Constructing a neural network model, and training the neural network model to obtain a breeding index evaluation model; The characteristic data is input into the breeding index evaluation model, and the poultry breeding index evaluation result is output.

2. The poultry breeding index evaluation method according to claim 1, characterized in that: The obtaining of characteristic data of poultry further comprises: Obtaining data on the physiological characteristics, egg-laying data, and hatching data of poultry; The eggs and hatching birds are marked separately and the data are linked to the laying birds; Clean the acquired data and remove unqualified data; Convert qualified data into feature data.

3. The poultry breeding index evaluation method according to claim 2, characterized in that: The step of obtaining the physiological characteristic data, egg laying data and hatching data of poultry further includes: The physiological characteristic data of the poultry include: weight, age, body shape and genetic data of the poultry; The egg-laying data include: egg-laying quantity and egg size, shape, and color data; The hatching data include: hatched poultry weight, feather color, toe number and beak shape data.

4. The poultry breeding index evaluation method according to claim 1, characterized in that: The step of constructing a neural network model and training the neural network model to obtain a breeding index evaluation model further includes: Determine the number of layers of the neural network, the number of neurons in each layer, and the activation function; Compile the model, specify the loss function, optimizer, and evaluation metric; The historical data are used as samples to train the model, adjust the model parameters, and finally obtain the breeding index evaluation model.

5. The poultry breeding index evaluation method according to claim 4, characterized in that: The method further includes: evaluating the breeding index evaluation model, further including: Setting a test set and a validation set, and inputting the test set into the breeding index evaluation model to obtain the prediction result of the model; Set and calculate evaluation indicators based on the prediction results; The obtained prediction results and evaluation indicators are compared with the validation set to evaluate the performance of the breeding indicator evaluation model.

6. The poultry breeding index evaluation method according to claim 1, characterized in that: The step of inputting the characteristic data into the breeding index evaluation model and outputting the poultry breeding index evaluation result further comprises: Inputting the characteristic data of the poultry to be tested into the breeding index evaluation model to obtain the probability that the poultry breeding index is qualified; Set a probability threshold, when the probability is greater than the threshold, it is qualified, and when the probability is less than the threshold, it is unqualified; Unqualified poultry are screened out based on the threshold and treated accordingly.

7. The poultry breeding index evaluation method according to claim 4, characterized in that: The method further comprises: The training samples are updated regularly, and the breeding index evaluation model parameters are updated according to the training samples.

8. A poultry breeding index evaluation device, characterized in that: include: A data acquisition module, used for acquiring characteristic data of poultry; A model training module is used to construct a neural network model and train the neural network model to obtain a breeding index evaluation model; The breeding index evaluation module is used to input the characteristic data into the breeding index evaluation model and output the poultry breeding index evaluation results.

9. An electronic device for evaluating poultry breeding indicators, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

10. A computer readable medium having computer program instructions stored thereon, characterized in that: The computer program instructions can be executed by a processor to implement the method according to any one of claims 1-7.

Images