Information processing apparatus, information processing method, and information processing program

Abstract

The learning device (100) has: an acquisition unit (120) that acquires learning data (10) including partial shape information (11) and PMI (12), the partial shape information (11) being information representing a set of elements of a shape of a machining site, the PMI (12) being information representing a machining content; an extraction unit (130) that extracts the partial shape information (11) and the PMI (12) from the learning data; and a generation unit (150) that generates a trained model using the partial shape information (11) and the PMI (12).

Description

Technical Field

[0001] This invention relates to information processing apparatus, information processing method, and information processing program. Background Technology

[0002] Manufacturing is carried out based on design data such as CAD (Computer-Aided Design) data. Additionally, the design data includes PMI (Product Manufacturing Information). Drawing inspection is performed on the design data. Here, a technique related to drawing inspection is proposed (see Patent Document 1).

[0003] Patent Document 1: Japanese Patent Application Publication No. 2022-105917 Summary of the Invention

[0004] When reviewing drawings, the accuracy of the PMI (Property Management Index) needs to be assessed. One approach to drawing review is to use rule information representing the rules. However, the accuracy of the PMI sometimes requires a comprehensive assessment. In such cases, a rule-based approach is not feasible. Therefore, the question arises as to how to make a comprehensive assessment of the PMI's accuracy.

[0005] The purpose of this invention is to comprehensively assess the accuracy of PMI.

[0006] An information processing apparatus according to one aspect of the present invention is provided. The information processing apparatus includes: an acquisition unit that acquires learning data comprising partial shape information and processing content information, wherein the partial shape information is information representing a set of elements indicating the shape of a processing part, and the processing content information is information representing the processing content; an extraction unit that extracts the partial shape information and the processing content information from the learning data; and a generation unit that generates a trained model using the partial shape information and the processing content information.

[0007] The effects of the invention

[0008] According to the present invention, a comprehensive judgment can be made on the accuracy of PMI. Attached Figure Description

[0009] Figure 1 This is a diagram illustrating the information processing system of Implementation Method 1.

[0010] Figure 2 This is a diagram showing the hardware of the learning device in Implementation 1.

[0011] Figure 3 This is a block diagram illustrating the functions of the learning device in Implementation 1.

[0012] Figure 4This is a diagram showing partial shape information and a specific example of PMI in Implementation Method 1.

[0013] Figure 5 This is a diagram showing an example of a conversion table for Implementation Method 1.

[0014] Figure 6 (A) and (B) are diagrams representing specific examples of the converted partial shape information and the converted PMI in Implementation 1.

[0015] Figure 7 This is a diagram illustrating a specific example of the integration process in Implementation Method 1.

[0016] Figure 8 This is a flowchart illustrating an example of the processing performed by the learning device in Implementation 1.

[0017] Figure 9 This is a block diagram illustrating the function of the estimation device in Embodiment 1.

[0018] Figure 10 This is a flowchart illustrating an example of the processing performed by the estimation device in Implementation 1.

[0019] Figure 11 This is a block diagram illustrating the function of the learning device in Embodiment 2.

[0020] Figure 12 This is a flowchart illustrating an example of the processing performed by the learning device in Implementation 2.

[0021] Figure 13 This is a block diagram illustrating the function of the estimation device in Embodiment 2.

[0022] Figure 14 This is a flowchart illustrating an example of the processing performed by the estimation device in Implementation 2.

[0023] Figure 15 This is a flowchart illustrating an example of the processing performed by the estimation device in Implementation 3. Detailed Implementation

[0024] Hereinafter, embodiments will be described with reference to the accompanying drawings. The embodiments described below are merely examples, and various modifications can be made within the scope of this invention.

[0025] Implementation Method 1

[0026] Figure 1This diagram illustrates an information processing system according to Embodiment 1. The information processing system includes a learning device 100 and an estimation device 200. The learning device 100 and the estimation device 200 may also be implemented using the same devices. The learning device 100 and the estimation device 200 may also be referred to as information processing devices. The learning device 100 is a device for executing an information processing method. This information processing method may also be referred to as a learning method. The estimation device 200 is a device for executing an information processing method. This information processing method may also be referred to as an estimation method.

[0027] Next, the hardware of the learning device 100 and the estimation device 200 will be described.

[0028] Figure 2 This diagram illustrates the hardware of the learning device according to Embodiment 1. The learning device 100 is a computer. The learning device 100 includes a processor 101, a volatile storage device 102, and a non-volatile storage device 103.

[0029] The processor 101 controls the learning device 100 as a whole. For example, the processor 101 is a CPU (Central Processing Unit), an FPGA (Field Programmable Gate Array), etc. The processor 101 can also be a multiprocessor. In addition, the learning device 100 may also have processing circuitry.

[0030] Volatile storage device 102 is the main storage device of learning device 100. For example, volatile storage device 102 is RAM (Random Access Memory). Non-volatile storage device 103 is an auxiliary storage device of learning device 100. For example, non-volatile storage device 103 is HDD (Hard Disk Drive) or SSD (Solid State Drive).

[0031] Like the learning device 100, the estimation device 200 has a processor, a volatile memory device, and a non-volatile memory device. Additionally, the estimation device 200 may also have processing circuitry.

[0032] The learning phase and the application phase will be described below. In the learning phase, the learning device 100 will be described. In the application phase, the estimation device 200 will be described.

[0033] <Learning Phase>

[0034] The functions of the learning device 100 will be explained.

[0035] Figure 3This is a block diagram illustrating the functions of the learning device according to Embodiment 1. The learning device 100 includes a storage unit 110, a retrieval unit 120, a retrieval unit 130, a conversion unit 140, and a generation unit 150. Furthermore, the conversion unit 140 may also be referred to as the first conversion unit.

[0036] The storage unit 110 may also be implemented as a storage area reserved in the volatile storage device 102 or the non-volatile storage device 103.

[0037] Part or all of the acquisition unit 120, extraction unit 130, conversion unit 140, and generation unit 150 can be implemented by a processing circuit. Alternatively, part or all of the acquisition unit 120, extraction unit 130, conversion unit 140, and generation unit 150 can also be implemented as modules of a program executed by the processor 101. For example, the program executed by the processor 101 is also called an information processing program or a learning program. For example, the information processing program or learning program is recorded on a recording medium.

[0038] Storage unit 110 stores various types of information. Storage unit 110 can also store the learning data 10 and the conversion table described later.

[0039] The acquisition unit 120 acquires learning data 10. For example, the acquisition unit 120 acquires learning data 10 from the storage unit 110. Alternatively, for example, the acquisition unit 120 acquires learning data 10 from an external device. Furthermore, the external device is a device located outside the learning device 100. For example, the external device is a cloud server, external storage, etc. A diagram of the external device is omitted.

[0040] Let's explain learning data 10. For example, learning data 10 is CAD data. Learning data 10 includes partial shape information 11 and PMI 12. Partial shape information 11 is information about the set of elements representing the shape of the processed part. Alternatively, partial shape information 11 can also be expressed as information about the set of elements representing the shape of the part that appears due to processing. For example, in the case of making a hole in a plate, the processed object is the plate, and partial shape information 11 is information about the set of elements representing the shape of the cylindrical hole. Specifically, partial shape information 11 is LINE, CURVE, SURFACE, EDGE, etc., representing the cylindrical hole (in other words, cylinder). Partial shape information 11 can also be represented by row numbers in learning data 10, as in the specific example shown later. PMI 12 is information representing the processing content. PMI 12 is also called processing content information. In addition, PMI 12 is correct data. That is, the information shown in PMI 12 is correct information as processing content. In other words, the information shown in PMI 12 is correct information after a comprehensive judgment of the processing content. For example, in the case of a hole in a board, PMI 12 is information related to the size of the hole. Specific examples of partial shape information 11 and PMI 12 are shown.

[0041] Figure 4 This is a diagram showing partial shape information and a specific example of PMI in Implementation Method 1. Figure 4 An example of learning data 10 is shown. Additionally, Figure 4 Examples of partial shape information 11 and PMI 12 are shown. Furthermore, Figure 4 Partial shape information 11 shows the row numbers of the elements of the shape of the "cylindrical hole".

[0042] return Figure 3 The extraction unit 130 will be described below.

[0043] The extraction unit 130 extracts partial shape information 11 and PMI 12 from the learning data 10. For example, the extraction unit 130 uses a table for determining the partial shape information 11 and PMI 12 to perform the extraction.

[0044] The conversion unit 140 converts at least one of the partial shape information 11 and PMI 12 into a predetermined form. Specifically, the conversion unit 140 uses a conversion table or a trained model (hereinafter, a conversion trained model) to convert the partial shape information 11 and PMI 12 into a predetermined form.

[0045] Here, the reason for the conversion is explained. When a user (e.g., a designer) creates learning data 10 (e.g., CAD data), the partial shape information 11 and PMI 12 contained in the learning data 10 are recorded by the user using various expressions. For example, partial shape information 11 and PMI 12 are expressed using color. For example, partial shape information 11 and PMI 12 are expressed using text. For example, partial shape information 11 and PMI 12 are expressed using symbols. Thus, partial shape information 11 and PMI 12 are recorded by the user using various expressions. Therefore, in order to unify the expression methods, the conversion unit 140 converts the partial shape information 11 and PMI 12.

[0046] First, the use of a conversion table by the conversion unit 140 will be explained. The conversion table is obtained by the acquisition unit 120. For example, the acquisition unit 120 obtains the conversion table from the storage unit 110. The acquisition unit 120 may also obtain the conversion table from an external device. An example of a conversion table is shown.

[0047] Figure 5 This is a diagram illustrating an example of a conversion table according to Embodiment 1. For example, conversion table 111 is stored in storage unit 110. Conversion table 111 may also be referred to as conversion information. Conversion table 111 is information used to convert partial shape information 11 and PMI 12 into a predetermined form. Conversion table 111 has items for description content and conversion content. Description content is registered in the description content item. Conversion content is registered in the conversion content item.

[0048] The conversion unit 140 uses conversion table 111 to convert partial shape information 11 and PMI 12 into a predetermined form. For example, the conversion unit 140 analyzes the partial shape information 11 and creates a tree structure representing the connection relationships of elements such as lines, surfaces, and curves. Furthermore, the analysis process is the same as the usual analysis process for CAD data. Additionally, the tree structure consists of multiple nodes. These nodes represent elements such as lines, surfaces, and curves. The conversion unit 140 uses conversion table 111 to convert elements such as lines into a predetermined form. The conversion unit 140 traverses the tree structure, arranging the elements of each node separated by spaces. This generates data in the form of statements. The conversion unit 140 can also convert to a predetermined form after arranging the elements of each node. The method for traversing the tree structure can be depth-first search, width-first search, etc. A specific example of the converted partial shape information 11 and PMI 12 is shown.

[0049] Figure 6 (A) and (B) are diagrams representing specific examples of the converted partial shape information and the converted PMI in Implementation 1.Figure 6 (A) shows an example of the converted partial shape information. Figure 6 (B) shows an example of the converted PMI.

[0050] Next, the case where the conversion unit 140 uses the conversion-trained model will be described. The conversion-trained model is acquired by the acquisition unit 120. For example, the acquisition unit 120 acquires the conversion-trained model from the storage unit 110. The acquisition unit 120 may also acquire the conversion-trained model from an external device.

[0051] The conversion unit 140 uses a conversion-trained model to convert partial shape information 11 and PMI 12 into a predetermined form. Specifically, when the conversion unit 140 inputs partial shape information 11 into the conversion-trained model, the conversion-trained model outputs partial shape information 11 expressed in a predetermined form (i.e., the converted partial shape information). Similarly, when the conversion unit 140 inputs PMI 12 into the conversion-trained model, the conversion-trained model outputs PMI 12 expressed in a predetermined form (i.e., the converted PMI).

[0052] In this way, partial shape information 11 and PMI 12 are converted into a predetermined form.

[0053] The generation unit 150 uses partial shape information and PMI to generate a trained model. Furthermore, when the partial shape information is transformed, the partial shape information used to generate the trained model is the transformed partial shape information. Similarly, when the PMI is transformed, the PMI used to generate the trained model is the transformed PMI.

[0054] A trained model, when given a PMI (Problem Detection Index), outputs information indicating whether that PMI is correct, or outputs the correct PMI when given a PMI. Furthermore, whether the PMI is correct refers to its accuracy as information for processing.

[0055] Alternatively, the generation unit 150 can be converted into a state that integrates the converted partial shape information and the converted PMI, using the converted information to generate a trained model. An example of the converted information is shown.

[0056] Figure 7 This is a diagram illustrating a specific example of the integration process in Implementation Method 1. Figure 7 This shows the state after the transformation of the partial shape information and the transformed PMI through integration (i.e., the transformed information).

[0057] Next, a flowchart will be used to explain the processes performed by the learning device 100.

[0058] Figure 8 This is a flowchart illustrating an example of the processing performed by the learning device in Implementation 1.

[0059] (Step S11) Obtain learning data 10 from the acquisition unit 120.

[0060] (Step S12) Extraction unit 130 extracts partial shape information 11 and PMI 12 from learning data 10.

[0061] (Step S13) The conversion unit 140 converts the partial shape information 11 and PMI 12 into a predetermined form.

[0062] (Step S14) The generation unit 150 uses the converted partial shape information and the converted PMI to learn the learning model.

[0063] (Step S15) The generation unit 150 determines whether the learning has ended. For example, if the acquisition unit 120 receives an end instruction input by the user, the learning ends. If the learning has ended, the learning model becomes a trained model. If the learning has not ended, the process proceeds to step S11.

[0064] The trained model is stored in the estimation device 200 or an external device.

[0065] In this way, a well-trained model is generated. The well-trained model is generated using PMI 12 as correct data; therefore, the learning method in the learning phase can also be called teacher-guided learning.

[0066] According to the learning phase, the learning device 100 learns by using learning data, thereby generating a well-trained model that comprehensively judges the accuracy of the PMI. The learning device 100 prepares this well-trained model. Thus, by preparing the well-trained model, it is possible to comprehensively judge the accuracy of the PMI.

[0067] The conversion of partial shape information 11 and PMI 12 has been explained above. When partial shape information 11 has already been expressed in a predetermined form, the conversion unit 140 does not convert partial shape information 11. Similarly, when PMI 12 has already been expressed in a predetermined form, the conversion unit 140 does not convert PMI 12. Without converting partial shape information 11 and PMI 12, the generation unit 150 generates a trained model using partial shape information 11 and PMI 12.

[0068] Next, the application phase will be explained.

[0069] <Application Phase>

[0070] The functions of the estimation device 200 are explained.

[0071] Figure 9 This is a block diagram illustrating the function of the estimation device according to Embodiment 1. The estimation device 200 includes a storage unit 210, an acquisition unit 220, an extraction unit 230, a conversion unit 240, an estimation unit 250, a comparison unit 260, and an output unit 270. Furthermore, the conversion unit 240 may also be referred to as the first conversion unit.

[0072] The storage unit 210 may also be implemented as a storage area reserved in the volatile or non-volatile storage device of the presumed device 200.

[0073] Part or all of the acquisition unit 220, extraction unit 230, conversion unit 240, estimation unit 250, comparison unit 260, and output unit 270 can be implemented by the processing circuitry of the estimation device 200. Alternatively, part or all of the acquisition unit 220, extraction unit 230, conversion unit 240, estimation unit 250, comparison unit 260, and output unit 270 can also be implemented as modules of a program executed by the processor of the estimation device 200. For example, the program executed by the processor is also called an information processing program or an estimation program. For example, the information processing program or estimation program is recorded on a recording medium.

[0074] Storage unit 210 stores various types of information. Storage unit 210 can also store CAD data 20 and conversion table 111, which will be described later.

[0075] The acquisition unit 220 acquires CAD data 20. For example, the acquisition unit 220 acquires CAD data 20 from the storage unit 210. Alternatively, for example, the acquisition unit 220 acquires CAD data 20 from an external device. Furthermore, CAD data 20 is also referred to as design data. CAD data 20 includes partial shape information 21 and PMI 22.

[0076] The acquisition unit 220 acquires the trained model from the storage unit 210 or an external device. Furthermore, this trained model is a trained model generated by the learning device 100.

[0077] Extraction unit 230 extracts partial shape information 21 and PMI 22 from CAD data 20. For example, extraction unit 230 uses a table for determining partial shape information 21 and PMI 22 for extraction.

[0078] The conversion unit 240 converts at least one of the partial shape information 21 and PMI 22 into a predetermined form. Specifically, the conversion unit 240 uses conversion table 111 or a conversion-trained model to convert the partial shape information 21 and PMI 22 into a predetermined form. The method of using conversion table 111 and conversion-trained model is the same as described above, so the description is omitted.

[0079] The estimation unit 250 uses partial shape information, PMI, and a trained model to output information indicating whether PMI 22 is correct, or to output the correct PMI. Furthermore, when the partial shape information is transformed, the partial shape information input to the trained model is the transformed partial shape information. When the PMI is transformed, the PMI input to the trained model is the transformed PMI. For example, when the estimation unit 250 inputs the transformed partial shape information and the transformed PMI to the trained model, the trained model outputs information indicating whether PMI 22 is correct, or the correct PMI.

[0080] Alternatively, the estimation unit 250 can be converted to a state that integrates the converted partial shape information and the converted PMI, and using the converted information and the trained model, output information indicating whether PMI 22 is correct, or output the correct PMI.

[0081] If the correct PMI is output, the comparison unit 260 compares PMI 22 or the converted PMI with the correct PMI. For example, the comparison unit 260 uses word2vec or BERT for comparison.

[0082] When information indicating whether PMI 22 is correct is output, the output unit 270 outputs that information. For example, the output unit 270 outputs the information to the display of the estimation device 200. By outputting this information, the designer can identify whether PMI 22 is correct.

[0083] After comparison, the output unit 270 outputs the comparison result. For example, the output unit 270 outputs the comparison result to the display of the estimation device 200. By outputting the comparison result, the designer can identify whether the PMI 22 is excessive or insufficient.

[0084] Next, a flowchart will be used to explain the processes performed by the estimation device 200.

[0085] Figure 10 This is a flowchart illustrating an example of the processing performed by the estimation device in Implementation 1. Furthermore, in the following description, the trained model outputs the correct PMI.

[0086] (Step S21) The acquisition unit 220 acquires CAD data 20.

[0087] (Step S22) Extraction unit 230 extracts partial shape information 21 and PMI 22 from CAD data 20.

[0088] (Step S23) The conversion unit 240 converts the partial shape information 21 and PMI 22 into a predetermined form.

[0089] (Step S24) The estimation unit 250 uses the converted partial shape information, the converted PMI and the trained model to output the correct PMI.

[0090] (Step S25) The comparison unit 260 compares the PMI 22 or the converted PMI with the correct PMI.

[0091] (Step S26) Output unit 270 outputs the comparison result.

[0092] Depending on the application stage, the estimation device 200 can make a comprehensive judgment on the accuracy of PMI by using a trained model.

[0093] The conversion of partial shape information 21 and PMI 22 has been described above. When partial shape information 21 is expressed in a predetermined form, the conversion unit 140 does not convert partial shape information 21. Similarly, when PMI 22 is expressed in a predetermined form, the conversion unit 140 does not convert PMI 22. When partial shape information 21 and PMI 22 are not converted, the estimation unit 250 uses partial shape information 21, PMI 22, and the trained model to output information indicating whether PMI 22 is correct, or to output the correct PMI.

[0094] Here, sometimes CAD data 20 does not include PMI 22. When PMI 22 is not included in CAD data 20, the estimation unit 250 inputs partial shape information 21 or converted partial shape information into the trained model. The trained model then outputs information indicating that PMI 22 is not included in CAD data 20. The output unit 270 outputs this information. By outputting this information, the designer can notice that PMI 22 is not included in CAD data 20.

[0095] Implementation Method 2

[0096] Next, Embodiment 2 will be described. In Embodiment 2, the differences from Embodiment 1 will be mainly explained. Furthermore, in Embodiment 2, the description of matters common to Embodiment 1 will be omitted.

[0097] <Learning Phase>

[0098] Figure 11 This is a block diagram illustrating the function of the learning device according to Embodiment 2. The learning device 100 also includes a conversion unit 160. The conversion unit 160 may also be referred to as a second conversion unit.

[0099] Part or all of the conversion unit 160 may be implemented by processing circuitry. Alternatively, part or all of the conversion unit 160 may also be implemented as a module of a program executed by the processor 101.

[0100] In Embodiment 2, the learning data 10 is in binary format. Since the extraction unit 130 cannot extract the data if it is not in ASCII format, the conversion unit 160 converts the learning data 10 to ASCII format. In other words, the conversion unit 160 converts the learning data 10 into a text file. For example, the conversion unit 160 uses a table for format conversion to convert the learning data 10 to ASCII format. With the learning data 10 converted to ASCII format, the extraction unit 130 can then extract it.

[0101] Next, a flowchart will be used to explain the processes performed by the learning device 100.

[0102] Figure 12 This is a flowchart illustrating an example of the processing performed by the learning device in Implementation 2. Figure 12 processing and Figure 8 The difference in processing lies in the execution of step S11a. Therefore, in Figure 12 The following section describes step S11a. Furthermore, descriptions of processes other than step S11a are omitted.

[0103] (Step S11a) The conversion unit 160 converts the format of the learning data 10 into ASCII format.

[0104] In addition, in step S12, the extraction unit 130 extracts partial shape information 11 and PMI 12 from the converted learning data 10.

[0105] According to Embodiment 2, the learning device 100 can perform learning even when the learning data 10 is in binary format.

[0106] <Application Phase>

[0107] Figure 13 This is a block diagram illustrating the function of the estimation device in Embodiment 2. The estimation device 200 also includes a conversion unit 280. The conversion unit 280 may also be referred to as a second conversion unit.

[0108] Part or all of the conversion unit 280 may be implemented by the processing circuitry of the estimation device 200. Alternatively, part or all of the conversion unit 280 may be implemented as a module of a program executed by the processor of the estimation device 200.

[0109] In Embodiment 2, the CAD data 20 is in binary format. Since the extraction unit 230 cannot extract the data if it is not in ASCII format, the conversion unit 280 converts the CAD data 20 to ASCII format. In other words, the conversion unit 280 converts the CAD data 20 to a text file. For example, the conversion unit 280 uses a table for format conversion to convert the CAD data 20 to ASCII format. With the CAD data 20 converted to ASCII format, the extraction unit 230 can then extract the data.

[0110] Next, a flowchart will be used to explain the processes performed by the estimation device 200.

[0111] Figure 14 This is a flowchart illustrating an example of the processing performed by the estimation device in Implementation 2. Figure 14 processing and Figure 10 The difference in processing lies in the execution of step S21a. Therefore, in Figure 14 The following section describes step S21a. Furthermore, descriptions of processes other than step S21a are omitted.

[0112] (Step S21a) The conversion unit 280 converts the format of the CAD data 20 into ASCII format.

[0113] In addition, in step S22, the extraction unit 230 extracts partial shape information 21 and PMI 22 from the converted CAD data 20.

[0114] According to Embodiment 2, the estimation device 200 can perform estimation even when the CAD data 20 is in binary format.

[0115] Implementation Method 3

[0116] Next, Embodiment 3 will be described. In Embodiment 3, the differences from Embodiment 1 will be mainly explained. Furthermore, in Embodiment 3, the descriptions of matters common to Embodiment 1 will be omitted.

[0117] <Application Phase>

[0118] In Embodiment 1, the extraction of all partial shape information 21 and PMI 22 contained in the CAD data 20 was described. In Embodiment 3, the extraction of elements, etc., corresponding to the PMI of the estimated object from the partial shape information 21 was described.

[0119] The processes performed by the estimation device 200 are described using a flowchart.

[0120] Figure 15 This is a flowchart illustrating an example of the processing performed by the estimation device in Implementation 3. Figure 15 processing and Figure 10 The difference in processing lies in the execution of steps S21b and 22a. Therefore, in Figure 15 The steps S21b and S22a will be explained in detail. Furthermore, the explanation of processes other than steps S21b and S22a will be omitted.

[0121] (Step S21b) The acquisition unit 220 acquires the estimated object information from the storage unit 210 or an external device. Furthermore, the estimated object information shows the shape elements that are related to the PMI that is the estimated object, and the PMI that is the estimated object.

[0122] (Step S22a) The extraction unit 230 uses the estimated object information to extract partial shape information 21 and the estimated object PMI 22 from the CAD data 20. The partial shape information 21 represents the shape elements that are related to the estimated object PMI.

[0123] According to Embodiment 3, the estimation device 200 does not extract all the partial shape information 21 and PMI 22 contained in the CAD data 20. Therefore, the time for performing the conversion process (i.e., step S23) and the estimation process (i.e., step S24) is short. Furthermore, the time for performing the estimation process is short when no conversion process is performed. As a result, the estimation device 200 can reduce its processing burden.

[0124] The features described above in the various embodiments can be appropriately combined with each other.

[0125] Explanation of the label

[0126] 10 Learning data, 11 Partial shape information, 12 PMI, 20 CAD data, 21 Partial shape information, 22 PMI, 100 Learning device, 101 Processor, 102 Volatile storage device, 103 Non-volatile storage device, 110 Storage unit, 111 Conversion table, 120 Acquisition unit, 130 Extraction unit, 140 Conversion unit, 150 Generation unit, 160 Conversion unit, 200 Estimation device, 210 Storage unit, 220 Acquisition unit, 230 Extraction unit, 240 Conversion unit, 250 Estimation unit, 260 Comparison unit, 270 Output unit, 280 Conversion unit.

Claims

1. An information processing device, comprising: The acquisition unit acquires learning data containing partial shape information and processing content information. The partial shape information is information about a set of elements representing the shape of the processing part, and the processing content information is information representing the processing content. The extraction unit extracts the partial shape information and the processing content information from the learning data; as well as The generation unit uses the partial shape information and the processing content information to generate a trained model.

2. The information processing apparatus according to claim 1, wherein, It also includes a first conversion unit that converts at least one of the partial shape information and the processing content information into a predetermined form. When the partial shape information has been transformed, the partial shape information used to generate the trained model is the transformed partial shape information. When the processed content information is transformed, the processed content information used to generate the trained model is the transformed processed content information.

3. The information processing apparatus according to claim 1 or 2, wherein, It also has a second conversion unit. The learning data is in binary format. The second conversion unit converts the learning data into ASCII format. The extraction unit extracts the partial shape information and the processing content information from the converted learning data.

4. An information processing device, comprising: The acquisition department acquires a trained model and design data containing partial shape information and processing content information. The partial shape information is information about a set of elements representing the shape of the processing part, and the processing content information is information about the processing content. The extraction unit extracts the partial shape information and the processing content information from the design data; as well as The estimation unit uses the partial shape information, the processing content information, and the trained model to output information indicating whether the processing content information is correct, or to output the correct processing content information.

5. The information processing apparatus according to claim 4, wherein, It also includes a first conversion unit that converts at least one of the partial shape information and the processing content information into a predetermined form. After the partial shape information has been transformed, the partial shape information input to the trained model is the transformed partial shape information. When the processed content information is transformed, the processed content information input to the trained model is the transformed processed content information.

6. The information processing apparatus according to claim 4 or 5, wherein, It also has: The comparison unit, when outputting correct processing content information, compares the processing content information or the converted processing content information with the correct processing content information; and The output section outputs the comparison results.

7. The information processing apparatus according to any one of claims 4 to 6, wherein, It also has a second conversion unit. The design data is in binary format. The second conversion unit converts the design data into ASCII format. The extraction unit extracts the partial shape information and the processing content information from the converted design data.

8. The information processing apparatus according to any one of claims 4 to 7, wherein, The acquiring unit acquires presumed object information, which represents shape elements that are related to the processing content information that is the presumed object, and the processing content information that is the presumed object. The extraction unit uses the estimated object information to extract from the design data the processing content information that is the estimated object and the partial shape information of the elements representing the shape that is related to the processing content information that is the estimated object.

9. The information processing apparatus according to claim 4, wherein, It also has an output section. The design data does not include the processing content information. The estimation unit inputs the partial shape information or the transformed partial shape information into the trained model. The output unit outputs the following information, which is output from the trained model, indicating that the processing content information is not included in the design data.

10. An information processing method, wherein, The information processing device acquires learning data containing partial shape information and processing content information. The partial shape information is a set of elements representing the shape of the processing part, and the processing content information is information representing the processing content. The information processing device extracts the partial shape information and the processing content information from the learning data. The information processing device uses the partial shape information and the processing content information to generate a trained model.

11. An information processing method, wherein, The information processing device acquires a trained model and design data containing partial shape information and processing content information. The shape information is a set of elements representing the shape of the processing part, and the processing content information is information representing the processing content. The information processing device extracts the partial shape information and the processing content information from the design data. The information processing device uses the partial shape information, the processing content information, and the trained model to output information indicating whether the processing content information is correct, or to output the correct processing content information.

12. An information processing program that causes an information processing device to perform the following processes, The learning data includes partial shape information and processing content information. The shape information is a set of elements representing the shape of the processing part, and the processing content information is information representing the processing content. Extract the partial shape information and the processing content information from the learning data. The trained model is generated using the partial shape information and the processing content information.

13. An information processing program that causes an information processing device to perform the following processes, Obtain a trained model and design data containing partial shape information and processing content information. The shape information is a set of elements representing the shape of the processing part, and the processing content information is information representing the processing content. Extract the partial shape information and the processing content information from the design data. Using the partial shape information, the processing content information, and the trained model, output information indicating whether the processing content information is correct, or output the correct processing content information.

Images