A cloud metering coding full-process material management system and method

Abstract

The present invention belongs to the technical field of material management. A cloud-based metering coding full-process material management system and method is provided. The system includes a quantity counting module, a material matching module, a price inquiry module, and an integrated management module. Compared with the method of manually counting and manually converting the quantity counting list into a price inquiry form, the scheme of the present invention can realize the automatic execution of quantity counting and price inquiry, significantly reduce labor costs, and because the materials are managed online by coding, the materials can be efficiently managed throughout the whole process.

Description

Technical Field

[0001] The present invention relates to the technical field of material management, and in particular to a cloud metering coding full-process material management system and method. Background Art

[0002] Before project implementation or project bidding, after obtaining the construction drawings, it is necessary to conduct a quantity statistics of all equipment and materials required for the project according to the construction drawings. It is necessary to count the specifications and quantities of different materials, and inquire about the prices of all equipment and materials to determine the approximate cost of the project.

[0003] The quantity counting is mainly done manually. After the counting is completed, the list is exported and the list is made into an inquiry form before the inquiry is made. This traditional material management method completely separates the counting and inquiry, which is not conducive to the subsequent tracking of the use of each material, especially the macro statistics of the price fluctuations of different types of materials. Moreover, manual counting and converting the counting list into an inquiry form will consume a lot of manpower and reduce efficiency. Summary of the invention

[0004] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a cloud metering coding whole-process material management method, system, electronic device, computer storage medium and computer program product.

[0005] The present invention provides a cloud metering and coding full-process material management system, the system comprising a point quantity module, a material matching module, a price inquiry module, and an integrated management module;

[0006] The point quantity module is used to analyze the first construction plan information, extract the required equipment, materials and their specifications and quantities, and form a first material quantity calculation table;

[0007] The material matching module is used to match each material in the first material calculation table with a preset material in a dictionary library, assign the code of the matched preset material to the corresponding material in the first material calculation table, and obtain a second material calculation table;

[0008] The inquiry module is used to upload the second material quantity calculation table to the inquiry system, obtain a feedback price list, match and integrate the price list with the second material quantity calculation table, and obtain a third material quantity calculation table;

[0009] The integrated management module is used to manage various materials across projects based on the codes to output material statistical data; and to receive user settings for total material plan and requests for adding missing materials.

[0010] The present invention also provides a cloud metering coding whole process material management method, based on the above system, the method includes the following method steps:

[0011] Receiving the input first construction plan information, the point quantity module analyzes the first construction plan information, extracts the required equipment, materials and their specifications and quantities therefrom, and forms a first material quantity calculation table;

[0012] The material matching module matches each material in the first material calculation table with a preset material in a dictionary library, assigns a code of the matched preset material to a corresponding material in the first material calculation table, and obtains a second material calculation table;

[0013] The inquiry module uploads the second material quantity calculation table to the inquiry system, obtains a feedback price list, matches and integrates the price list with the second material quantity calculation table, and obtains a third material quantity calculation table;

[0014] The integrated management module manages each material across projects based on the code to output material statistical data; and receives the user's setting of the total material plan and the request for adding the missing materials.

[0015] In some embodiments, the point quantity module analyzes the first construction plan information, extracts the required equipment, materials, their specifications and quantities, and forms a first material quantity calculation table, including:

[0016] The point quantity module calls the semantic analysis model to perform semantic analysis on the first construction plan information, and extracts several groups of material information, wherein the material information includes material attribute information, specification information, and quantity information;

[0017] The point quantity module calls the automatic point quantity software to perform point quantity calculation on each of the material information to obtain the first calculated quantity material table.

[0018] In some embodiments, the semantic analysis model is trained in the following manner:

[0019] Collect and construct a small sample training data set; wherein the small sample training data set includes several groups of training data, each group of the training data includes the second construction plan information and the fourth material quantity calculation table belonging to the same other project;

[0020] Each piece of the training data is input into the BERT large model to perform fine-tuning training on the BERT large model. After the fine-tuning training meets the termination condition, the semantic analysis model is obtained.

[0021] In some embodiments, inputting each piece of the training data into the BERT large model to perform fine-tuning training on the BERT large model includes:

[0022] Calculating the total amount of the training data contained in the small sample training data set, and determining a number of prompting opportunities according to the total amount;

[0023] Input the training data one by one into the BERT large model for fine-tuning training, and suspend the fine-tuning training when the prompt timing is reached;

[0024] Randomly select a piece of the training data that has not been input, input the second construction plan information contained therein into the BERT large model for testing, and obtain a fifth material quantity calculation table, calculate the deviation between the fifth material quantity calculation table and the fourth material quantity calculation table corresponding to the second construction plan information, and correct the next prompting opportunity according to the deviation;

[0025] Generate prompt information according to the fifth material calculation table, the fourth material calculation table and the degree of deviation, wherein the prompt information is used to prompt a user to input guidance information for guiding the training of the BERT large model;

[0026] End the pause and input the guidance information and the uninput training data into the BERT large model one by one to continue fine-tuning training.

[0027] In some embodiments, the modifying the next prompting opportunity according to the deviation includes:

[0028] An adjustment value is calculated based on the deviation, and the adjustment value is used to adjust the input cumulative value of the training data corresponding to the next prompt timing.

[0029] In some embodiments, the price inquiry module uploads the second calculated material quantity list to the price inquiry system to obtain a feedback price list, including:

[0030] The inquiry module uploads the second material quantity calculation table and the expected construction duration corresponding to the first construction plan information to the inquiry system;

[0031] The inquiry system determines the number of inquiry channels according to the expected construction time, screens and obtains a number of inquiry ports according to the number of inquiry channels, sends the second material quantity calculation table to each of the inquiry ports, obtains the corresponding first price list, and integrates the first price lists to obtain the second price list;

[0032] The inquiry module receives the second price list fed back by the inquiry system.

[0033] The beneficial effect of the present invention is that compared with the method of manually counting quantities and manually converting the count list into an inquiry form, the above-mentioned scheme of the present invention can realize the automatic execution of count and inquiry, significantly reduce labor costs, and because of the use of coding to manage materials online, the materials can be efficiently managed throughout the entire process.

[0034] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0036] Figure 1 It is a structural diagram of a cloud metering coding full-process material management system disclosed in an embodiment of the present invention;

[0037] Figure 2 It is a schematic diagram of the principle of a cloud metering coding whole-process material management method disclosed in an embodiment of the present invention;

[0038] Figure 3 It is a flow chart of a cloud metering coding whole-process material management method disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0040] See also Figure 1 , the embodiment of the present invention discloses a cloud metering coding whole process material management system, the system includes a point quantity module, a material matching module, a price inquiry module, and an integrated management module;

[0041] The point quantity module is used to analyze the first construction plan information, extract the required equipment, materials and their specifications and quantities, and form a first material quantity calculation table;

[0042] The material matching module is used to match each material in the first material calculation table with a preset material in a dictionary library, assign the code of the matched preset material to the corresponding material in the first material calculation table, and obtain a second material calculation table;

[0043] The inquiry module is used to upload the second material quantity calculation table to the inquiry system, obtain a feedback price list, match and integrate the price list with the second material quantity calculation table, and obtain a third material quantity calculation table;

[0044] The integrated management module is used to manage various materials across projects based on the codes to output material statistical data; and to receive user settings for total material plan and requests for adding missing materials.

[0045] Compared with the existing inefficient method of manual counting and price inquiry, the present invention designs the above-mentioned cloud metering coding whole process material management system, which includes a counting module, a material matching module, a price inquiry module, and an integrated management module. Among them, the counting module can automatically analyze the first construction plan information, extract the equipment, materials and their specifications and quantities required for the corresponding construction project, and form a first quantity calculation material table. The material matching module matches each material in the first quantity calculation material table with the preset materials in the dictionary library one by one, so as to obtain the corresponding code of the matched preset materials, so as to obtain the second quantity calculation material table with the configuration material code. The price inquiry module automatically uploads the second quantity calculation material table to the price inquiry system, and the price inquiry system queries the market price of each material, matches and integrates the feedback price list with the aforementioned second quantity calculation material table, and further obtains the third quantity calculation material table, which at least includes the equipment / material name, specification, quantity, code, and price parameters. Finally, the integrated management module uses the above coding as a benchmark to achieve cross-project management of various materials. That is, after entering a material code, you can retrieve the price of the material in different projects, the price fluctuations in different periods, etc. The use of materials can be effectively tracked through coding. In addition, the integrated management module can also support users (such as engineers) to perform operations such as total material planning, and when there are missing materials in the calculated material library, users can also add the name, specification, quantity and other data of the missing materials to achieve material addition. The added materials will be synchronized to the calculated material table in the software material library, and then complete the coding configuration and inquiry in the above manner.

[0046] Compared with the method of manually counting quantities and manually converting the count list into an inquiry form, the above-mentioned scheme of the present invention can realize the automatic execution of count and inquiry, significantly reduce labor costs, and because the materials are managed online by coding, the materials can be efficiently managed throughout the whole process.

[0047] The dictionary library is preset, and the user pre-arranges and codes the materials involved in various types of construction projects to construct the above dictionary library. In addition, the material management system of the present invention is arranged in the cloud.

[0048] See also Figure 2, Figure 3 The embodiment of the present invention discloses a cloud metering coding whole process material management method. Based on the above system, the method includes the following steps:

[0049] Receiving the input first construction plan information, the point quantity module analyzes the first construction plan information, extracts the required equipment, materials and their specifications and quantities therefrom, and forms a first material quantity calculation table;

[0050] The material matching module matches each material in the first material calculation table with a preset material in a dictionary library, assigns a code of the matched preset material to a corresponding material in the first material calculation table, and obtains a second material calculation table;

[0051] The inquiry module uploads the second material quantity calculation table to the inquiry system, obtains a feedback price list, matches and integrates the price list with the second material quantity calculation table, and obtains a third material quantity calculation table;

[0052] The integrated management module manages each material across projects based on the code to output material statistical data; and receives the user's setting of the total material plan and the request for adding the missing materials.

[0053] Since the method in this embodiment adopts all the technical solutions of the above-mentioned system embodiment, it has at least all the beneficial effects brought by the technical solutions of the above-mentioned embodiment, which will not be described one by one here.

[0054] In some embodiments, the point quantity module analyzes the first construction plan information, extracts the required equipment, materials, their specifications and quantities, and forms a first material quantity calculation table, including:

[0055] The point quantity module calls the semantic analysis model to perform semantic analysis on the first construction plan information, and extracts several groups of material information, wherein the material information includes material attribute information, specification information, and quantity information;

[0056] The point quantity module calls the automatic point quantity software to perform point quantity calculation on each of the material information to obtain the first calculated quantity material table.

[0057] In this embodiment, the present invention configures a semantic analysis model and automatic point quantity software for the point quantity module. The point quantity module can configure corresponding analysis tasks for the two by calling, thereby simplifying the size of the point quantity module.

[0058] The user inputs the first construction plan information of this time. Since the first construction plan information contains many words that are irrelevant to equipment, materials, their specifications and quantities, the automatic counting software cannot analyze these irrelevant words. Therefore, the present invention sets the counting module to first call the semantic analysis model to perform semantic analysis on the first construction plan information, and extract several groups of material information. The material information includes material attribute information (mainly equipment / material name), specification information, and quantity information, so that the material information that does not contain or only contains a small amount of irrelevant words is obtained from the first construction plan information. Then call the automatic counting software to perform point quantity calculation on each material information, and the first calculation material table can be obtained. Of course, after the first calculation material table is generated, it can also be output to the user for confirmation and modification, thereby further improving the accuracy of the first calculation material table.

[0059] Among them, the automatic measurement software can be the Glodon BIM civil engineering measurement platform (GTJ version), engineering material cloud EMCloud, Beam AI, etc., and the present invention does not make specific limitations on this.

[0060] In some embodiments, the semantic analysis model is trained in the following manner:

[0061] Collect and construct a small sample training data set; wherein the small sample training data set includes several groups of training data, each group of the training data includes the second construction plan information and the fourth material quantity calculation table belonging to the same other project;

[0062] Each piece of the training data is input into the BERT large model to perform fine-tuning training on the BERT large model. After the fine-tuning training meets the termination condition, the semantic analysis model is obtained.

[0063] In this embodiment, the present invention uses the BERT big model to construct the above-mentioned semantic analysis model, so that the existing BERT big model with powerful comprehensive analysis capabilities can be used to reduce the difficulty and cost of building a dedicated vertical model alone. Specifically, the second construction plan information of other projects and the fourth quantity material table actually generated are first collected, wherein the second construction plan information constitutes a training master data, and the fourth quantity material table constitutes the training prompt data. Therefore, under the "prompt" of the fourth quantity material table, the general BERT big model can gradually train the accurate recognition of the material information in the training master data, thereby removing those irrelevant words, and finally obtaining a semantic analysis model of the first quantity material table that can accurately analyze the material information.

[0064] In some embodiments, inputting each piece of the training data into the BERT large model to perform fine-tuning training on the BERT large model includes:

[0065] Calculating the total amount of the training data contained in the small sample training data set, and determining a number of prompting opportunities according to the total amount;

[0066] Input the training data one by one into the BERT large model for fine-tuning training, and suspend the fine-tuning training when the prompt timing is reached;

[0067] Randomly select a piece of the training data that has not been input, input the second construction plan information contained therein into the BERT large model for testing, and obtain a fifth material quantity calculation table, calculate the deviation between the fifth material quantity calculation table and the fourth material quantity calculation table corresponding to the second construction plan information, and correct the next prompting opportunity according to the deviation;

[0068] Generate prompt information according to the fifth material calculation table, the fourth material calculation table and the degree of deviation, wherein the prompt information is used to prompt a user to input guidance information for guiding the training of the BERT large model;

[0069] End the pause and input the guidance information and the uninput training data into the BERT large model one by one to continue fine-tuning training.

[0070] In this embodiment, the present invention improves the traditional fine-tuning training method by adding a visualization stage during the fine-tuning training process. The user can analyze the flaws of the previous fine-tuning training based on the output visualization information (i.e., the above-mentioned prompt information), and then input guidance information. The guidance information will be input into the BERT large model together with other training data that have not been input, so as to implement subsequent fine-tuning training under the guidance of the guidance information.

[0071] Specifically, a number of prompting opportunities are first determined based on the total amount of training data contained in the small sample training data set. The prompting opportunities can be evenly arranged, for example, if the total amount of training data is 100, the moments when 25, 50, and 75 pieces of training data are input can be determined as prompting opportunities; or they can be unevenly arranged, for example, the moments when 25, 60, and 85 pieces of training data are input can be determined as prompting opportunities. The present invention does not make specific limitations on this.

[0072] When a prompting opportunity is reached, fine-tuning training is suspended, and a piece of training data that has not been input into the BERT large model for fine-tuning training is randomly selected as test data, and the second construction plan information contained therein is input into the BERT large model for testing. The BERT large model that has undergone partial fine-tuning training will output the corresponding fifth material calculation table. Then, the deviation between the analyzed fifth material calculation table and the corresponding true value, i.e., the fourth material calculation table, is calculated and analyzed, and the next prompting opportunity can be corrected according to the deviation. Among them, the deviation can be the ratio of the number of materials in the fifth material calculation table that are not recorded in the fourth material calculation table to the number of materials in the fourth material calculation table.

[0073] For this prompt, the fifth material quantity calculation table, the fourth material quantity calculation table corresponding to the test data, and the calculated deviation are directly filled into the prompt information. The user can analyze the flaws of the previous fine-tuning training according to the prompt information, and then enter the guidance information, such as "For the parking lot construction part of the construction plan information, the XX equipment involved is the same as the XX equipment involved in the outdoor road, which can be reused without repeated measurement". Then, resume the fine-tuning training, input the guidance information and the uninput training data one by one into the BERT large model to continue fine-tuning training, and the BERT large model can adjust its measurement strategy based on the above guidance information.

[0074] The above-mentioned guidance information of the present invention can also be input together with the training data at the beginning of fine-tuning training, but the guidance information is actually difficult to determine in advance, because the specific location of the training defects of the BERT large model cannot be predicted in advance, so by pausing and inputting guidance information, this defect can be effectively overcome.

[0075] In some embodiments, the modifying the next prompting opportunity according to the deviation includes:

[0076] An adjustment value is calculated based on the deviation, and the adjustment value is used to adjust the input cumulative value of the training data corresponding to the next prompt timing.

[0077] In this embodiment, the prompting time is determined by the input cumulative value of the training data, such as the aforementioned 25, 50, and 75. In this regard, the present invention converts the deviation obtained above to obtain the corresponding adjustment value, and then uses the adjustment value to adjust the input cumulative value of the training data corresponding to the next prompting time in a targeted manner, thereby realizing the correction of the next prompting time.

[0078] For example, if the deviation is 30%, the adjustment value is converted to 0.8. Assuming that the cumulative input value of the training data corresponding to the next prompt opportunity is 50, the adjustment value multiplied by the cumulative input value is equal to 50*0.8=40, that is, the prompt information is output in advance when 40 training data are cumulatively input.

[0079] Among them, there should be a negative correlation between the deviation and the adjustment value, that is, the larger the deviation, the poorer the current fine-tuning training effect, and the amount of single training data should be reduced, and the training direction of the BERT large model should be intervened by introducing guidance information in advance, so as to obtain a better fine-tuning training effect. However, the present invention does not specifically limit the conversion formula between the deviation and the adjustment value.

[0080] In some embodiments, the price inquiry module uploads the second calculated material quantity list to the price inquiry system to obtain a feedback price list, including:

[0081] The inquiry module uploads the second material quantity calculation table and the expected construction duration corresponding to the first construction plan information to the inquiry system;

[0082] The inquiry system determines the number of inquiry channels according to the expected construction time, screens and obtains a number of inquiry ports according to the number of inquiry channels, sends the second material quantity calculation table to each of the inquiry ports, obtains the corresponding first price list, and integrates the first price lists to obtain the second price list;

[0083] The inquiry module receives the second price list fed back by the inquiry system.

[0084] In this embodiment, different construction projects require different construction times, and the prices of materials used in construction will change over time. The longer the construction time, the greater the probability and amplitude of such changes. In response to this problem, the present invention sets an inquiry module to upload the aforementioned second material calculation table and the expected construction time corresponding to the first construction plan information to the inquiry system. The inquiry system then determines the number of inquiry channels based on the expected construction time, and selects several inquiry ports based on the number of inquiry channels. For example, if the number of inquiry channels is 5, the top 5 inquiry ports are selected from the preset 10 inquiry ports. The second material calculation table is sent to each inquiry port for inquiry, so as to obtain the corresponding 5 first price tables, and the above-mentioned second price table can be obtained by merging and calculating these first price tables (for example, calculating the average value). Compared with the first price table, the second price table is calculated based on multi-source prices, and is less likely to be affected by price fluctuations, so that the reference of the material prices in the obtained second price table can be ensured to a certain extent, which is convenient for users to perform engineering cost analysis and other purposes.

[0085] Among them, the inquiry port refers to the data port of different material quotation service providers, such as material quotation websites or APPs.

[0086] The present invention also provides an electronic device, comprising: a memory storing executable program code; and a processor coupled to the memory; characterized in that: the processor calls the executable program code stored in the memory to execute the method described in the above embodiment.

[0087] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0088] The present invention further provides a computer storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method described in the above embodiment is executed.

[0089] In the context of the present disclosure, a computer storage medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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 of the foregoing.

[0090] The present invention also provides a computer program product, including a computer program stored on a non-transitory computer-readable medium, wherein the computer program implements the method described in the above embodiment when executed by a processor.

[0091] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0092] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0093] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A cloud metering coding full-process material management method, based on a cloud metering coding full-process material management system, characterized in that: The system includes a point quantity module, a material matching module, a price inquiry module, and an integrated management module; The point quantity module is used to analyze the first construction plan information, extract the required equipment, materials and their specifications and quantities, and form a first material quantity calculation table; The material matching module is used to match each material in the first material calculation table with a preset material in a dictionary library, assign the code of the matched preset material to the corresponding material in the first material calculation table, and obtain a second material calculation table; The inquiry module is used to upload the second material quantity calculation table to the inquiry system, obtain a feedback price list, match and integrate the price list with the second material quantity calculation table, and obtain a third material quantity calculation table; The integrated management module is used to manage various materials across projects based on the codes to output material statistics; and receive user settings for total material plan and requests for adding missing materials; The method comprises the following steps: Receiving the input first construction plan information, the point quantity module analyzes the first construction plan information, extracts the required equipment, materials and their specifications and quantities therefrom, and forms a first material quantity calculation table; The material matching module matches each material in the first material calculation table with a preset material in a dictionary library, assigns a code of the matched preset material to a corresponding material in the first material calculation table, and obtains a second material calculation table; The inquiry module uploads the second material quantity calculation table to the inquiry system, obtains a feedback price list, matches and integrates the price list with the second material quantity calculation table, and obtains a third material quantity calculation table; The integrated management module manages each material across projects based on the code to output material statistics; and receives the user's setting of the total material plan and the request for adding the missing materials; The point quantity module analyzes the first construction plan information, extracts the required equipment, materials, their specifications and quantities, and forms a first material quantity calculation table, including: The point quantity module calls the semantic analysis model to perform semantic analysis on the first construction plan information, and extracts several groups of material information, wherein the material information includes material attribute information, specification information, and quantity information; The point quantity module calls the automatic point quantity software to perform point quantity calculation on each of the material information to obtain the first calculated material table; The semantic analysis model is trained in the following way: Collect and construct a small sample training data set; wherein the small sample training data set includes several groups of training data, each group of the training data includes the second construction plan information and the fourth material quantity calculation table belonging to other projects; Input each piece of the training data into the BERT large model to perform fine-tuning training on the BERT large model, and obtain the semantic analysis model after the fine-tuning training meets the termination condition; Inputting each piece of the training data into the BERT large model to perform fine-tuning training on the BERT large model, including: Calculating the total amount of the training data contained in the small sample training data set, and determining a number of prompting opportunities according to the total amount; Input the training data one by one into the BERT large model for fine-tuning training, and suspend the fine-tuning training when the prompt timing is reached; Randomly select a piece of the training data that has not been input, input the second construction plan information contained therein into the BERT large model for testing, and obtain a fifth material quantity calculation table, calculate the deviation between the fifth material quantity calculation table and the fourth material quantity calculation table corresponding to the second construction plan information, and correct the next prompting opportunity according to the deviation; wherein the deviation is the ratio of the number of materials in the fifth material quantity calculation table that are not recorded in the fourth material quantity calculation table to the number of materials in the fourth material quantity calculation table; Generate prompt information according to the fifth material calculation table, the fourth material calculation table and the degree of deviation, wherein the prompt information is used to prompt a user to input guidance information for guiding the training of the BERT large model; End the pause, and input the guidance information and each of the training data that has not been input into the BERT large model one by one to continue fine-tuning training; Correcting the next prompting timing according to the deviation includes: An adjustment value is calculated based on the deviation, and the adjustment value is used to adjust the input cumulative value of the training data corresponding to the next prompt timing; wherein the deviation and the adjustment value are negatively correlated.

2. According to claim 1, a cloud metering coding whole process material management method is characterized by: The inquiry module uploads the second calculated material quantity table to the inquiry system to obtain a feedback price table, including: The inquiry module uploads the second material quantity calculation table and the expected construction duration corresponding to the first construction plan information to the inquiry system; The inquiry system determines the number of inquiry channels according to the expected construction time, screens and obtains a number of inquiry ports according to the number of inquiry channels, sends the second material quantity calculation table to each of the inquiry ports, obtains the corresponding first price list, and integrates the first price lists to obtain the second price list; The inquiry module receives the second price list fed back by the inquiry system.

Images