Building space visual comfort multi-dimensional intelligent evaluation method based on model knowledge migration and computer system

Abstract

The invention discloses a model knowledge migration-based building space visual comfort multi-dimensional intelligent evaluation method, which comprises the following steps of: forming a target data set by collecting an existing image evaluation open source data set and newly added data according to an annotation rule, and carrying out standardization and normalization processing on an evaluation label by utilizing a fitting normal distribution form; a multi-gating hybrid expert model is applied, pre-training and distillation learning are carried out through a teacher-student model, and multi-dimensional comprehensive evaluation of image colors, light rays and architectural layout is realized through migration of model knowledge and updating of normal distribution parameters. The method has the advantages that the accuracy and interpretability of evaluation are improved, the integration of subjective and objective evaluation and the efficient application of knowledge migration are also realized, and an intelligent and comprehensive evaluation tool is provided for the visual comfort design of the building space.

Description

Technical Field

[0001] The present invention relates to the field of architectural design and comfort evaluation, and in particular to a multi-dimensional intelligent evaluation method and computer system for architectural space visual comfort based on model knowledge migration. Background Art

[0002] With the continuous development of intelligent technology and the deepening of the people-oriented concept in architectural space design, the quality requirements for public building spaces are gradually increasing. Visual comfort, as one of the key factors affecting the quality of public building space design, is directly related to the physiological and mental health in daily life. Traditional visual comfort evaluation is usually based on the evaluation of a single dimension of physical parameters such as quantitative light intensity and color temperature, which is difficult to fully reflect the subjective visual experience of users in complex architectural spaces. At the same time, due to the diversity of spatial functions and the variability of user needs, the evaluation method based on a small amount of objective data through sampling methods such as questionnaires in the past cannot adapt to the visual comfort needs of different scenarios.

[0003] With the continuous advancement of intelligent technology, in the field of computer vision and image evaluation, image-based intelligent evaluation technology for building scene quality continues to develop. Subjective scoring and scoring algorithms for different dimensions of images, such as color, light, and layout, are becoming more and more widely used. More and more image evaluation datasets are open source, but evaluation datasets of different dimensions often have problems such as incomplete labels and inconsistent label dimensions, making it difficult to uniformly apply them to the visual comfort evaluation of building scene images. The main problems are:

[0004] 1) The difficulty of migrating image datasets with different evaluation dimensions and building unified label datasets: Existing data such as lighting, color, and layout come from various sources and are often built for a single evaluation dimension or for different evaluation purposes, with heterogeneous characteristics. How to pre-train based on existing image datasets, efficiently integrate existing open source data through knowledge migration, normalize different evaluation labels, and introduce new datasets are important challenges in building evaluation models.

[0005] 2) Explainability of the multi-dimensional intelligent evaluation algorithm for visual comfort: In the past, some evaluation algorithms often evaluated a single dimension, and set some prompt words, such as "Which scene looks more comfortable", to let users select a picture to achieve the classification of comfort levels. However, the evaluation of visual comfort is obviously affected by multiple dimensions, and how to conduct a comprehensive evaluation of multiple dimensions is another challenge.

[0006] This is an area where this application needs to be improved. Summary of the invention

[0007] The technical problem to be solved by the present invention is to provide a multi-dimensional intelligent evaluation method and computer system for the visual comfort of architectural space based on model knowledge transfer, aiming to improve the quality of architectural space design and meet the comfort requirements of modern public spaces. It not only improves the accuracy and interpretability of the evaluation, but also realizes the fusion of subjective and objective evaluation and the efficient application of knowledge transfer, providing an intelligent and comprehensive evaluation tool for the visual comfort design of architectural space.

[0008] In order to solve the above technical problems, the present invention provides a multi-dimensional intelligent evaluation method for architectural space visual comfort based on model knowledge migration, comprising the following steps:

[0009] Step S1: Establishment of image multi-dimensional evaluation dataset;

[0010] Collect existing open-source image evaluation datasets, including image color dimension evaluation datasets, image light evaluation datasets, and evaluation datasets of different dimensions of building layouts, and continuously supplement the building space pictures taken on site to expand the dataset;

[0011] Step S2: label division of the newly added data set;

[0012] Extract two different architectural space pictures from the dataset to be evaluated at one time. According to the prompt word: "Which of the following pictures performs better in terms of color / light / architectural layout, etc.", the evaluation of the corresponding dimension of the selected picture will receive corresponding rewards and punishments. After multiple rounds of scoring, the label of the current dataset is finally formed;

[0013] By fitting a normal distribution To reflect the evaluation distribution of the current building picture, where μ i Indicates the score of the current image in a certain evaluation dimension. The larger the value, the higher the score. Indicates the stability of the current rating. The smaller the value, the more stable the evaluation of the current image. The larger the value, the more unstable and controversial the evaluation of the object.

[0014] Step S3: normalize the labels of the migration dataset;

[0015] Since existing evaluation datasets are often evaluated on different dimensions, the label designs are also different. Some are based on fitting normal distribution or other probability distribution forms, while others are directly divided into 1, 2, 3...n levels. In order to migrate datasets with different labels to the same model, these datasets are allowed to target only one dimension, but their labels must be normalized to the normal distribution. On; Similarly, μ i Indicates the score of the current image in a certain evaluation dimension, corresponding to the original label evaluation value of the migration dataset; Indicates the stability of the current score. However, since the migration dataset may not have this label, a default value is set based on the dataset size.

[0016] Step S4: pre-training the multi-gated hybrid expert model using the migration dataset;

[0017] Pre-training the multi-gated hybrid expert model using the normalized migration dataset in step S3;

[0018] A multi-gated mixed expert model is used to fit the μ of the input image in each dimension i , σ i ; The framework of the multi-gated hybrid expert model includes the following aspects: first, after the data is input, it passes through an attention mechanism layer, and then enters a gating network and several expert layers respectively. Among them, each expert layer is composed of the encoder module of the Transformer. Each expert layer receives the same information from the attention mechanism layer, and then the output parameters are weighted and combined by the weight values ​​output by the gating network; the gating network is composed of several layers of neural networks, and the terminal activation function selects softmax. The weighted reorganized data is then input into each discriminant layer of the corresponding evaluation dimension respectively. The discriminant layer is also composed of several layers of neural networks. Finally, each discriminant layer outputs the evaluation value μ of the corresponding dimension respectively. i ,σ i ;

[0019] The basic loss value of the multi-gated hybrid expert model is composed of the KL divergence of the model output and input distribution. Since there are multiple evaluation dimensions, the weighted average is taken as the final basic loss value;

[0020] Loss ini =∑ i＝1 w i D KL (In i |Out i ),∑ i＝1 w i =1;

[0021] Step S5: The multi-gated hybrid expert model learns the newly added data set and realizes the multi-dimensional evaluation of visual comfort;

[0022] The setting of the student model is the same as the multi-gated hybrid expert model structure described in step S4. Because the scene objects described by the migration dataset are not necessarily all architectural spaces, but are pure color combination pictures or some pictures that can reflect the lighting effect, the migration dataset often cannot be directly trained with the newly added dataset. However, the migration dataset has guiding value for the evaluation dimensions of color, lighting, layout, etc. of the pictures.

[0023] Use the pre-trained model in step S4 as the teacher model, train a new multi-gated hybrid expert model of the same framework as the student model to learn the newly added data set labeled in step S2, input the newly added data set into the teacher model and the student model, and obtain the output μ on different evaluation dimensions. T , and μ S , And calculate the KL divergence D between the two sets of outputs KL (T|S), used to measure the distillation loss of the teacher model and the student model;

[0024]

[0025] This part of the loss value is weighted with the loss value set by the original model to coordinate the training of the model, realize the migration of multivariate data and the multi-dimensional evaluation of visual comfort, and the distillation loss Loss between the input teacher model and the output of the student model is calculated. dist Multiplying by the scaling factor α, the final overall loss value is:

[0026] Loss total =Loss ini +αLoss dist .

[0027] The scaling factor α is set according to the initial values ​​of the two loss values ​​and is dynamically adjusted as the model is further trained. The purpose is to keep the two loss values ​​of the same order of magnitude and avoid serious bias and oscillation during training.

[0028]

[0029] The step S2 comprises the following specific steps:

[0030] Step S21: firstly initialize the labels of the dataset to be evaluated;

[0031] Set the evaluation value μ of all the data to be evaluated in the dimension to be tested (color / light / building layout) o ∈

[0032] [0,100], initial μ i =50, Each dimension has a corresponding μ and σ;

[0033] Step S22: Set the probability P of two images being extracted for comparison when evaluating a certain dimension. AB , to ensure the stability of expected value and variance updates as much as possible;

[0034]

[0035]

[0036] Where: β is the variance of each comparison, which is adjusted according to the actual situation and is set by default c is used to measure the overall uncertainty caused by the two pictures in this comparison process. At the same time, in order to avoid a picture being evaluated too many times while some pictures have never been evaluated, it is necessary to set an upper limit on the number of times a single user can evaluate a picture. Pictures that reach this upper limit will no longer be extracted by the user.

[0037] Step S23: Each time two images are compared, a comparison prompt word on one dimension is obtained. After answering the comparison prompt word, basic reward and punishment values ​​R are obtained for the two images respectively. For example, a prompt word is obtained: "Which of the following images performs better in terms of color / light / architectural layout, etc.", which means that a comparison is only performed on one dimension. There are five options, namely A is better, B is better, about the same, both are good, and neither is good. The basic reward and punishment values ​​R brought by the five options to the two images are shown in Table 1:

[0038] Table 1 Basic reward value settings

[0039] Options <![CDATA[Reward and punishment value R A > <![CDATA[Reward and punishment value R B > A is better <![CDATA[R A >0]]> <![CDATA[R B >0]]> B is better <![CDATA[R A <0]]> <![CDATA[R B <0]]> almost <![CDATA[R A =0]]> <![CDATA[R B =0]]> All good <![CDATA[R A >0]]> <![CDATA[R B >0 <!-- 3 -->]]> Neither is good <![CDATA[R A <0]]> <![CDATA[R B <0]]>

[0040] Since two images theoretically have different distributions after multiple rounds of comparison, assuming that the evaluation in the first few rounds is valid, theoretically the image with a larger μ should win. However, there are always unexpected situations in the actual comparison process. Therefore, in addition to the basic reward and punishment values, we should also consider whether the comparison between the two images after several rounds of accumulation meets expectations. The specific expected value and standard deviation updates are shown in Table 2:

[0041] Table 2 Expected value update settings

[0042]

[0043]

[0044] Table 3 Standard deviation update settings

[0045]

[0046] in, In order to measure the overall uncertainty caused by the two images in this comparison process, ε is used as the threshold for judging the probability of a tie, which is preferably 0.5. The v and w functions in Table 3 are defined as follows:

[0047]

[0048] w(x,ε)=(x,ε)·(v(x,ε)+x-ε)

[0049] Where: φ(·) is the probability density function of the standard normal distribution, Φ(·) is the cumulative distribution function of the standard normal distribution;

[0050] Step S24:

[0051] Let different users repeatedly evaluate the data set to be tested, iterate the expected value and standard deviation of each image in the current image data set in different dimensions, and finally form the evaluation label of the data set.

[0052] The step S3 comprises the following specific steps:

[0053] Step S31: normalizing the data labels annotated in a normal distribution form;

[0054] For image evaluation labels using the normal distribution method standard, linear scaling is directly used to convert the original range to the current range. The specific formula is as follows:

[0055]

[0056] μ nnw ∈[0,max onw ],μ transfer ∈[0,max transfer ]

[0057] Step S32: normalizing the data labels using multi-level annotation;

[0058] For data labels that use a discrete grading system, they are mapped to a normal distribution based on the true meaning of their grading. For example, if a data set has the following grading definitions: 1, 2, 3, 4, 5, 1 is the worst and 5 is the best, then the mapping rules are as follows:

[0059]

[0060] The standard deviation is set to the default value. Or fine-tune according to the specific distribution of each level of the original data set: assuming that the evaluation of high-level and low-level images is stable, multiply the high-level images by a scaling factor; assuming that the evaluation of intermediate-level images is fuzzy and unstable, multiply this type of image by a larger scaling factor than the former; the specific decision is based on the size of the migration data set and the label distribution.

[0061] The present invention also provides a computer system, including a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the multi-dimensional intelligent evaluation method for architectural space visual comfort based on model knowledge migration.

[0062] The beneficial effects of the present invention are:

[0063] 1) Introducing a multi-dimensional evaluation dataset to improve the measurement dimension of architectural space visual comfort image evaluation;

[0064] By introducing multivariate data sets for model pre-training and knowledge transfer, we can help establish and improve the multi-dimensional measurement of architectural space visual comfort images;

[0065] 2) Achieve model knowledge transfer and alignment of evaluation labels to match the integration of different evaluation rules and standards;

[0066] 3) The teacher model is trained using the migration dataset, and the target student model is guided to learn the newly added target dataset through knowledge distillation to achieve the migration of model knowledge;

[0067] 4) The present invention proposes a label normalization method under different evaluation rule systems, which realizes the integration of multivariate evaluation data sets;

[0068] 5) The present invention uses a multi-gated hybrid expert model to output evaluation values ​​of the building space image to be tested in different dimensions, which more specifically highlights the aspects in which the visual comfort of the building space to be tested is better and where it is insufficient, which is more conducive to designers to optimize and design the solution and realize multi-dimensional intelligent evaluation of the visual comfort of the building space. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0070] Figure 1 It is a schematic diagram of a flow chart of a specific embodiment of the present invention;

[0071] Figure 2 A schematic diagram of a process for pre-training a multi-gated hybrid expert model according to a specific embodiment of the present invention;

[0072] Figure 3 A schematic diagram of an expert scoring process for label division of a newly added data set according to a specific embodiment of the present invention;

[0073] Figure 4 A schematic diagram of a multi-gated hybrid expert model learning a new data set according to a specific embodiment of the present invention;

[0074] Numbers in the figure:

[0075] S1—Establishment of image multi-dimensional evaluation dataset;

[0076] S2—label division of newly added data sets;

[0077] S3—label normalization of the migrated dataset;

[0078] S4—Pre-training the multi-gated mixture of experts model;

[0079] S5—New dataset for learning multi-gated mixture of experts model. DETAILED DESCRIPTION

[0080] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0081] like Figure 1 As shown, the present invention provides a multi-dimensional intelligent evaluation method for architectural space visual comfort based on model knowledge transfer, which forms a target data set by collecting existing image evaluation open source data sets and newly added data according to annotation rules, standardizes and normalizes evaluation labels in the form of fitting normal distribution, uses a multi-gated hybrid expert model, performs pre-training and distillation learning through a teacher-student model, and realizes multi-dimensional comprehensive evaluation of image color, light and architectural layout through the transfer of model knowledge and the update of normal distribution parameters; the method comprises the following steps:

[0082] Step S1: Establish an image multi-dimensional evaluation dataset;

[0083] Collect existing open source image evaluation datasets, including image evaluation data of color, light, and architectural layout dimensions to form a migration dataset, such as open source datasets such as AVA, place pulse, etc.; and take new pictures in the building space to add a new target dataset for a more comprehensive evaluation of the visual comfort of the building space;

[0084] In terms of specific operations, it is necessary to integrate various types of image evaluation datasets, which are not necessarily all image evaluation datasets related to the visual comfort of architectural images, or image datasets related to the color, lighting, etc. of the image itself or the evaluation of the architectural layout; then, according to the evaluation dimensions covered by the collected datasets, such as color, light and architectural layout, the newly added image data is classified and statistical labels are calculated.

[0085] Step S2: The target data set is divided into labels, such as Figure 3 As shown;

[0086] Generate labels for new datasets by comparing two images together;

[0087] Randomly select two pictures and evaluate them on a single dimension, such as color, light, or layout. You need to choose one of the options "A\B is better", "Similar", "Both are good", or "Both are bad". Each comparison will give the picture a reward or punishment value based on the user's choice, which is used to update the expected value and standard deviation of the label. After multiple evaluations, a normal distribution of the picture labels is established to ensure the stability and consistency of the evaluation. The specific steps are as follows:

[0088] Step S21: firstly initialize the labels of the dataset to be evaluated;

[0089] Set the evaluation value μ of all the data to be evaluated in the dimension to be tested (color / light / building layout) i ∈[0,100], initial μ i =50, Each dimension has a corresponding μ and σ;

[0090] Step S22: Set the probability P of two images being extracted for comparison when evaluating a certain dimension. AB , to ensure the stability of expected value and variance updates as much as possible;

[0091]

[0092] Where: β is the variance of each comparison, which is adjusted according to the actual situation and is set by default c is used to measure the overall uncertainty caused by the two pictures in this comparison process. At the same time, in order to avoid a picture being evaluated too many times while some pictures have never been evaluated, it is necessary to set an upper limit on the number of times a single user can evaluate a picture. Pictures that reach this upper limit will no longer be extracted by the user.

[0093] Step S23: Each time two images are compared, a prompt word will be obtained: "Which of the following images performs better in terms of color / light / architectural layout, etc." This means that a comparison is only performed on one dimension. There are five options, namely A is better, B is better, about the same, both are good, and neither is good. The five options are the basic reward and punishment values ​​R brought by the two images as shown in Table 1:

[0094] Table 1 Basic reward value settings

[0095] Options <![CDATA[Reward and punishment value R A > <![CDATA[Reward and punishment value R B > A is better <![CDATA[R A >0]]> <![CDATA[R B >0]]> B is better <![CDATA[R A <0]]> <![CDATA[R B <0]]> almost <![CDATA[R A =0]]> <![CDATA[R B =0]]> All good <![CDATA[R A >0]]> <![CDATA[R B >0]]> Neither is good <![CDATA[R A <0]]> <![CDATA[R B <0]]>

[0096] Since two images theoretically have different distributions after multiple rounds of comparison, assuming that the evaluation in the first few rounds is valid, theoretically the image with a larger μ should win. However, there are always unexpected situations in the actual comparison process. Therefore, in addition to the basic reward and punishment values, we should also consider whether the comparison between the two images after several rounds of accumulation meets expectations. The specific expected value and standard deviation updates are shown in Table 2:

[0097] Table 2 Expected value update settings

[0098]

[0099]

[0100] Table 3 Standard deviation update settings

[0101]

[0102] in, In order to measure the overall uncertainty caused by the two images in this comparison process, ε is used as the threshold for judging the probability of a tie, which is preferably 0.5. The v and w functions in Table 3 are defined as follows:

[0103]

[0104] w(x,ε)=(x,ε)·(v(x,ε)+x-ε)

[0105] Where: φ(·) is the probability density function of the standard normal distribution, Φ(·) is the cumulative distribution function of the standard normal distribution;

[0106] Step S24:

[0107] Let different users repeatedly evaluate the data set to be tested, iterate the expected value and standard deviation of each image in the current image data set in different dimensions, and finally form the evaluation label of the data set.

[0108] Step S3: normalize the labels of the migration dataset;

[0109] Since the labels of datasets from different sources are different, they need to be unified into the same distribution form. For data labels annotated in the form of normal distribution, they are converted to the range of the current model in a linear scale; for datasets with multi-level ratings, such as a rating system from 1 to 5, they are linearly mapped to a normal distribution to ensure the comparability of different labels. The specific steps are as follows:

[0110] Step S31: normalizing the data labels annotated in a normal distribution form;

[0111] For image evaluation labels using the normal distribution method standard, linear scaling is directly used to convert the original range to the current range. The specific formula is as follows:

[0112]

[0113] μ nnw ∈[0,max nnw ],μ transfer ∈[0,max transfer ]

[0114] Step S32: normalizing the data labels using multi-level annotation;

[0115] For data labels that use a discrete grading system, they are mapped to a normal distribution based on the true meaning of their grading. For example, if a data set has the following grading definitions: 1, 2, 3, 4, 5, 1 is the worst and 5 is the best, then the mapping rules are as follows:

[0116]

[0117] The standard deviation is set to the default value. Or fine-tune according to the specific distribution of each level of the original data set: assuming that the evaluation of high-level and low-level images is stable, multiply the high-level images by a scaling factor; assuming that the evaluation of intermediate-level images is fuzzy and unstable, multiply this type of image by a larger scaling factor than the former; the specific decision is based on the size of the migration data set and the label distribution.

[0118] Step S4: pre-training of teacher model of multi-gated hybrid expert model;

[0119] As attached Figure 2 As shown, the present invention adopts a multi-gated hybrid expert model to fit the μ of the input image in each dimension. i ,σ i . The Transformer-based multi-gated hybrid expert model is preliminarily trained using the migration dataset to lay the foundation for the evaluation of the new dataset. The model framework includes an attention mechanism layer, a gating network, an expert layer, and a discriminant layer to ensure the model's multi-dimensional input processing capabilities. The pre-training process is completed by inputting the normalized migration dataset into the model and calculating the weighted KL divergence loss function to optimize the model output;

[0120] The basic loss value of the multi-gated hybrid expert model is composed of the KL divergence of the model output and input distribution. Since there are multiple evaluation dimensions, the weighted average is taken as the final basic loss value;

[0121] Loss ini =∑ i＝1 w i D KL (In i |Out i ),∑ i＝1 w i =1;

[0122] Step S5: Student model training of multi-gated hybrid expert model, such as Figure 4 As shown;

[0123] A teacher-student model architecture is adopted for further optimization.

[0124] After the teacher model is pre-trained with the migration data set, it guides the student model to learn the new data set. The new data set is input into the teacher and student models respectively to obtain the output of each evaluation dimension. The difference between the teacher model and the student model is evaluated by calculating the KL divergence between the two sets of outputs, and the distillation loss is weightedly combined with the original loss value to optimize and update the model; the details are as follows:

[0125] Use the pre-trained model in step S4 as the teacher model, train a new multi-gated hybrid expert model of the same framework as the student model to learn the newly added data set labeled in step S2, input the newly added data set into the teacher model and the student model, and obtain the output μ on different evaluation dimensions. T , and μ S , And calculate the KL divergence D between the two sets of outputs KL (T|S), used to measure the distillation loss of the teacher model and the student model;

[0126]

[0127] This part of the loss value is weighted with the loss value set by the original model to coordinate the training of the model, realize the migration of multivariate data and the multi-dimensional evaluation of visual comfort, and the distillation loss Loss between the input teacher model and the output of the student model is calculated. dist Multiplying by the scaling factor α, the final overall loss value is:

[0128] Loss total =Loss ini +αLoss dist .

[0129] The present invention also provides a computer system, including a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the multi-dimensional intelligent evaluation method for architectural space visual comfort based on model knowledge migration.

[0130] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A multi-dimensional intelligent evaluation method for architectural space visual comfort based on model knowledge transfer, comprising the following steps: Step S1: Establishment of image multi-dimensional evaluation dataset; Collect existing open-source image evaluation datasets, including image color dimension evaluation datasets, image light evaluation datasets, and evaluation datasets of different dimensions of building layouts, and continuously supplement the building space pictures taken on site to expand the dataset; Step S2: label division of the newly added data set; Extract two different architectural space images from the dataset to be evaluated at one time, evaluate multiple dimensions of the selected images based on the prompt words and obtain corresponding rewards and punishments. After multiple rounds of scoring, the label of the current dataset is finally formed. By fitting a normal distribution To reflect the evaluation distribution of the current building picture, where μ i Indicates the score of the current image in a certain evaluation dimension. The larger the value, the higher the score. Indicates the stability of the current rating. The smaller the value, the more stable the evaluation of the current image. Step S3: normalize the labels of the migration dataset; Labels are normalized to a normal distribution Up; μ i Indicates the score of the current image in a certain evaluation dimension, corresponding to the original label evaluation value of the migration dataset; Indicates the stability of the current score; Step S4: pre-training the multi-gated hybrid expert model using the migration dataset; The framework of the multi-gated hybrid expert model includes an attention mechanism layer, a gating network, an expert layer, and a discriminant layer to fit the μ of the input image in each dimension. i , σ i ; The pre-training process is completed by inputting the normalized migration dataset into the model and calculating the weighted KL divergence loss function to optimize the model output; Step S5: The multi-gated hybrid expert model learns the newly added data set and realizes the multi-dimensional evaluation of visual comfort; The teacher-student model architecture is used for further optimization. After the teacher model is pre-trained with the migration dataset, it guides the student model to learn the new dataset. The new dataset is input into the teacher and student models respectively to obtain the output of each evaluation dimension. The difference between the teacher model and the student model is evaluated by calculating the KL divergence between the two sets of outputs, and the distillation loss is weightedly combined with the original loss value to optimize and update the model.

2. The multi-dimensional intelligent evaluation method for architectural space visual comfort based on model knowledge transfer according to claim 1 is characterized by: The step S2 comprises the following specific steps: Step S21: firstly initialize the labels of the dataset to be evaluated; Set the evaluation value μ of all the data to be evaluated on the dimension to be tested i ∈[0,100], initial μ i =50, Each dimension has a corresponding μ and σ; Step S22: Set the probability P of two images being extracted for comparison when evaluating a certain dimension. AB , ensuring the stability of expected value and variance updates; Where: β is the variance of each comparison, c is the overall uncertainty caused by the two pictures in this comparison process; Step S23: Each time two images are compared, a comparison prompt word on one dimension is obtained, and after answering the comparison prompt word, a basic reward or punishment value R is obtained for each of the two images; After multiple rounds of comparison, the two images have different distributions. Consider whether the comparison between the two images after several rounds of accumulation meets expectations. The specific expected value and standard deviation are updated as shown in Table 2: Table 2 Expected value update settings Table 3 Standard deviation update settings ε is used as the threshold for judging the probability of a draw, and the v and w functions in Table 3 are defined as follows: w(x,ε)=(x,ε)·(v(x,ε)+x-ε) Where: φ(·) is the probability density function of the standard normal distribution, Φ(·) is the cumulative distribution function of the standard normal distribution; Step S24: Let different users repeatedly evaluate the data set to be tested, iterate the expected value and standard deviation of each image in the current image data set in different dimensions, and finally form the evaluation label of the data set.

3. The multi-dimensional intelligent evaluation method for architectural space visual comfort based on model knowledge transfer according to claim 1 is characterized by: The step S3 comprises the following specific steps: Step S31: normalizing the data labels annotated in a normal distribution form; For image evaluation labels using the normal distribution method standard, linear scaling is directly used to convert the original range to the current range. The specific formula is as follows: m now ∈[0,max now ],m transfer ∈[0,max transfer ] Step S32: normalizing the data labels using multi-level annotation; For data labels that adopt discrete classification system, they are mapped to normal distribution according to the true meaning of their classification. The mapping rules are as follows:

4. The multi-dimensional intelligent evaluation method for architectural space visual comfort based on model knowledge transfer according to claim 1 is characterized by: The multi-gated hybrid expert model is that the data is first input through an attention mechanism layer, and then respectively input into a gating network and several expert layers, wherein each expert layer is composed of the encoder module of the Transformer, and each expert layer receives the same information from the attention mechanism layer, and then the output parameters are weightedly combined through the weight values ​​output by the gating network; the gating network is composed of several layers of neural networks, and the terminal activation function selects softmax. The weighted reorganized data is then respectively input into each discriminant layer corresponding to the evaluation dimension, and the discriminant layer is also composed of several layers of neural networks. Finally, each discriminant layer outputs the evaluation value μ of the corresponding dimension respectively. i ,σ i ; The basic loss value of the multi-gated hybrid expert model is composed of the KL divergence of the model output and input distribution. Since there are multiple evaluation dimensions, the weighted average is taken as the final basic loss value; Loss ini =∑ i＝1 w i D KL (In i |Out i ),∑ i＝1 w i =1。 5. The multi-dimensional intelligent evaluation method for architectural space visual comfort based on model knowledge transfer according to claim 1 is characterized by: The new data set is input into the teacher model and the student model to obtain the output μ on different evaluation dimensions. T , and μ S , And calculate the KL divergence D between the two sets of outputs KL (T|S), used to measure the distillation loss of the teacher model and the student model; The distillation loss Loss between the input teacher model and the output of the student model dist Multiplying by the scaling factor α, the final overall loss value is: Loss total =Loss ini +αLoss dist 。 6. A computer system comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the multi-dimensional intelligent evaluation method for architectural space visual comfort based on model knowledge migration as described in any one of claims 1 to 5.

Images